Feat/genui

[Devasy]

Summary by CodeRabbit

• New Features

  • Added AI-generated dashboards with charts, gauges, statistics, lists, and filter chips in Coach conversations.
  • Added health analytics for sleep, workout correlations, and muscle-group volume.
  • Added exercise handle selection and assisted-bodyweight tracking, including improved volume calculations.
  • Added handle-specific workout history and personal records.
  • Improved progression recommendations using recent sessions and deload recovery.

• Bug Fixes

  • Improved AI response reliability with retries, quota detection, model fallback, and Markdown fallback for unavailable dashboards.

[coderabbitai]

Walkthrough

Changes

A2UI dashboard rendering

Layer / File(s)	Summary
A2UI contracts, parser, and renderer workout-logger/lib/genui/*	Adds typed A2UI nodes, properties, registries, themes, panels, parsing, prompt generation, and rendering contracts.
Dashboard components workout-logger/lib/genui/src/components/*	Adds grid, list, chart, gauge, filter, radar, scatter, and statistic components.
Coach integration and validation workout-logger/lib/screens/ai_coach_screen.dart, workout-logger/lib/services/gemini_context_builder.dart, workout-logger/test/genui/*, workout-logger/test/screens/ai_coach_genui_test.dart	Coach messages now render A2UI payloads with Markdown fallback. Tests cover parsing, rendering, streaming, themes, registries, and malformed input.

Workout and AI features

Layer / File(s)	Summary
Handle and assisted-bodyweight data workout-logger/lib/models/models.dart, workout-logger/lib/data/exercise_database.dart, workout-logger/lib/services/settings_provider.dart	Adds handle choices, assistance and extra weight, assisted-bodyweight volume calculations, and persisted user body weight.
Workout history and recommendations workout-logger/lib/screens/widgets/exercise_input_section.dart, workout-logger/lib/screens/workout_flow_screen.dart, workout-logger/lib/services/workout_provider.dart, workout-logger/lib/services/managers/pr_manager.dart, workout-logger/lib/services/ml_service.dart	Adds handle selectors, handle-specific records and history, and deload-aware recommendations.
Health analytics and Gemini updates workout-logger/lib/services/ai/coach_tool_service.dart, workout-logger/lib/services/ai/gemini_ai_service.dart, workout-logger/lib/services/interfaces/*, workout-logger/scripts/test_gemini_api.py	Adds health analytics tools, chart-ready results, model fallback and retry handling, generic chat streaming, and structured JSON generation.

Build and repository configuration

Layer / File(s)	Summary
Native linker configuration .github/workflows/release.yml, fdroid/metadata/com.devasy.repforge.yml	Disables native linker build IDs in Flutter and F-Droid builds.
Repository metadata and design specification .gitignore, workout-logger/pubspec.yaml, docs/superpowers/specs/2026-08-08-sqlite-migration-and-coach-sql-tool-design.md	Ignores the .superpowers/ workspace, updates the package build number, and documents a future SQLite migration and SQL coach tool.

Possibly related PRs

• Devasy/RepForge#59: Updates related Android and Flutter build-toolchain configuration.

🚥 Pre-merge checks | ✅ 4 | ❌ 1

❌ Failed checks (1 warning)

Check name	Status	Explanation	Resolution
Docstring Coverage	⚠️ Warning	Docstring coverage is 37.50% which is insufficient. The required threshold is 80.00%.	Write docstrings for the functions missing them to satisfy the coverage threshold.

✅ Passed checks (4 passed)

Check name	Status	Explanation
Description Check	✅ Passed	Check skipped - CodeRabbit’s high-level summary is enabled.
Title check	✅ Passed	The title identifies the main GenUI feature introduced by the pull request and is concise, although it omits secondary workout and build changes.
Linked Issues check	✅ Passed	Check skipped because no linked issues were found for this pull request.
Out of Scope Changes check	✅ Passed	Check skipped because no linked issues were found for this pull request.

---

Thanks for using CodeRabbit! It's free for OSS, and your support helps us grow. If you like it, consider giving us a shout-out.

❤️ Share

• X
• Mastodon
• Reddit
• LinkedIn

_{Comment @coderabbitai help to get the list of available commands.}

[codecov]

Codecov Report

❌ Patch coverage is 83.62989% with 230 lines in your changes missing coverage. Please review.
✅ Project coverage is 75.79%. Comparing base (39fbb5e) to head (0ff6c2d).

Files with missing lines	Patch %	Lines
...out-logger/lib/services/ai/coach_tool_service.dart	70.61%	72 Missing ⚠️
...kout-logger/lib/services/ai/gemini_ai_service.dart	16.47%	71 Missing ⚠️
...er/lib/screens/widgets/exercise_input_section.dart	19.04%	34 Missing ⚠️
workout-logger/lib/services/workout_provider.dart	62.50%	15 Missing ⚠️
workout-logger/lib/services/settings_provider.dart	41.66%	7 Missing ⚠️
...-logger/lib/genui/src/components/metric_gauge.dart	94.11%	5 Missing ⚠️
workout-logger/lib/models/models.dart	86.11%	5 Missing ⚠️
...orkout-logger/lib/screens/workout_flow_screen.dart	64.28%	5 Missing ⚠️
.../lib/services/interfaces/ai_service_interface.dart	0.00%	4 Missing ⚠️
workout-logger/lib/genui/src/a2ui_prompt.dart	88.88%	3 Missing ⚠️
... and 7 more

Additional details and impacted files

@@            Coverage Diff             @@
##           r2.1.0      #64      +/-   ##
==========================================
+ Coverage   74.90%   75.79%   +0.88%     
==========================================
  Files          88      108      +20     
  Lines       14491    15839    +1348     
==========================================
+ Hits        10855    12005    +1150     
- Misses       3636     3834     +198     

☔ View full report in Codecov by Harness.
📢 Have feedback on the report? Share it here.

🚀 New features to boost your workflow:

• ❄️ Test Analytics: Detect flaky tests, report on failures, and find test suite problems.

[Devasy]

@coderabbitai

[coderabbitai]

@Devasy, I will review the changes in #64.

⚠️ Action not completed

Review rate limited.

Note: CodeRabbit is an incremental review system and does not re-review already reviewed commits. This command is applicable only when automatic reviews are paused.

[Devasy]

@coderabbitai review

[coderabbitai]

✅ Action performed

Review finished.

Note: CodeRabbit is an incremental review system and does not re-review already reviewed commits. This command is applicable only when automatic reviews are paused.

[coderabbitai]

Actionable comments posted: 48

🤖 Prompt for all review comments with AI agents

Verify each finding against current code. Fix only still-valid issues, skip the
rest with a brief reason, keep changes minimal, and validate.

Inline comments:
In @.github/workflows/release.yml:
- Around line 49-51: Update the linker-patching step after flutter pub get to
require and quote PUB_CACHE, restrict find to resolved jni-*/src/CMakeLists.txt
targets, and make the replacement idempotent so existing --build-id=none flags
are not duplicated. Remove the failure suppression and ensure the step exits
nonzero when no target file is found or patching fails.

In `@workout-logger/lib/genui/src/a2ui_panels.dart`:
- Around line 46-65: Update the trailing label Text in the Row alongside the
title to be wrapped with Flexible, and configure it with maxLines: 1 and
TextOverflow.ellipsis so model-provided labels cannot cause a RenderFlex
overflow.

In `@workout-logger/lib/genui/src/a2ui_parser.dart`:
- Around line 208-237: Update _extractJson to first attempt jsonDecode on the
complete stripped text and immediately return the decoded Map or List when
successful. Only run the existing balanced-candidate scan when whole-text
decoding fails, preserving the current fallback behavior for prose-wrapped JSON.
- Around line 78-98: Update the effective-props construction in the parser
around A2UiProps.stringKeyed and _parseChildren so that when json['props'] is a
Map, literal child-related keys from the outer json object are copied into
effective only when those keys are absent from props. Preserve props values when
both levels define the same key, and keep the existing flat-payload behavior
unchanged.

In `@workout-logger/lib/genui/src/a2ui_renderer.dart`:
- Around line 26-35: Update A2UiRenderer.build around the specFor lookup to emit
a debug-only diagnostic containing node.name when spec is null, then preserve
the existing SizedBox.shrink() fallback. Use the project’s existing debug
logging mechanism rather than changing rendering behavior.

In `@workout-logger/lib/genui/src/a2ui_spec.dart`:
- Around line 54-58: Update the A2UiSpec contract methods buildWidget and render
to use named parameters for all three arguments, then update every
implementation and call site consistently, including the renderer and parseProps
flow. Preserve the existing behavior and argument types, or document an explicit
exemption if the Widget build-style API must remain positional.

In `@workout-logger/lib/genui/src/a2ui_theme.dart`:
- Around line 41-42: Guard the public A2UiTheme palette contract so seriesColor
never evaluates a modulo operation with an empty seriesPalette. Update the
A2UiTheme constructor to assert or otherwise enforce a non-empty palette, while
preserving the existing indexed cycling behavior for valid palettes.

In `@workout-logger/lib/genui/src/components/dynamic_chart.dart`:
- Around line 223-249: Update _pie to filter props.series.first.values to
positive entries before calculating total or creating PieChartSectionData,
preserving each entry’s original index so props.labels remains aligned. Return
A2UiEmptyPanel when no positive values remain, and use the filtered values for
percentages and section geometry.

In `@workout-logger/lib/genui/src/components/metric_gauge.dart`:
- Around line 217-221: Update _GaugeArcPainter.shouldRepaint to also compare the
track property from oldDelegate, ensuring the painter repaints when the
background arc color changes while preserving the existing progress, from, and
to comparisons.

In `@workout-logger/lib/genui/src/components/stat_card.dart`:
- Around line 85-94: Update the value-formatting logic around rawValue, unit,
and the contains check to determine whether the unit is already present only
when it matches the trimmed suffix of rawValue. Preserve the existing fallback,
empty-unit handling, and spacing behavior while preventing short units from
matching unrelated text.

In `@workout-logger/lib/main.dart`:
- Around line 141-143: Update the CoachToolService constructor and every call
site, including the shown dependency injection call, to use named parameters for
all three dependencies. Preserve the existing dependency wiring while making
each argument explicit by name.

In `@workout-logger/lib/models/models.dart`:
- Around line 138-151: The calculateVolume method should use final for locals
that are not reassigned: replace the mutable effW declaration and conditional
assignments with a single final conditional expression, and change the drops
iteration variable from var drop to final drop.
- Around line 138-157: Persist assisted-bodyweight semantics so default volume
calculations use effective load rather than assistance weight: update
WorkoutSet.calculateVolume and its persisted fields in
workout-logger/lib/models/models.dart:138-157, aggregate the persisted
assisted-set calculation in ExerciseLog.totalVolume at
workout-logger/lib/models/models.dart:240-243, and update the logging flow at
workout-logger/lib/screens/workout_flow_screen.dart:548-555 to store assistance,
extra load, and a body-weight snapshot (or the effective load) for assisted
exercises.

In `@workout-logger/lib/screens/widgets/exercise_input_section.dart`:
- Around line 75-76: Update the assisted-load display calculations around
isAssistedBW and the equivalent values at the referenced later section to pass
every displayed weight through settings.toDisplay, including effectiveWeight,
settings.userBodyWeight, and currentWeight, while keeping the stored kilogram
values unchanged.
- Around line 75-76: The assisted-exercise classification is inconsistent
between the load panel and _InputRow for dips and push_ups. Reuse the existing
isAssistedBW result when constructing _InputRow, pass it into the widget, and
update _InputRow to use that value for its labeling instead of maintaining a
separate exercise-ID predicate.
- Around line 194-223: The handle selector in build must not display an
unpersisted first handle as selected; require or persist an explicit selection
before logging. In
workout-logger/lib/screens/widgets/exercise_input_section.dart:194-223, update
the active-selection logic accordingly. In
workout-logger/lib/services/workout_provider.dart:506-515, update
setExerciseHandle so existing recorded set handles are never rewritten; lock the
handle after the first set or create a separate log for each variation.

In `@workout-logger/lib/screens/workout_flow_screen.dart`:
- Around line 322-325: Make _loadLastSessionData pass the current log handle
from the onHandleChanged callback. In
workout-logger/lib/services/workout_provider.dart lines 674-696, require an
exact handle match whenever handle is non-empty, excluding logs with null or
different handles. Apply the same rule in lines 699-713 for last-session lookup,
using any legacy fallback only after no exact match exists.

In `@workout-logger/lib/services/ai/coach_tool_service.dart`:
- Around line 414-425: Clamp the model-provided days value in the tool method
before the loop that calls hh.sleepNight, reusing the existing _limitArg helper
and its supported range as other tools do. Keep the default of 14 for missing
input, then iterate using the clamped value.
- Around line 523-546: Update
workout-logger/lib/services/ai/coach_tool_service.dart#L523-L546 so
_getHealthMetrics uses the requested days window rather than always querying one
week, and align the get_health_metrics declaration with the sleep fields
actually returned instead of promising resting HR or readiness data. Update
workout-logger/lib/services/ai/coach_tool_service.dart#L599-L606 by removing the
resting_hr x_metric branch/description unless it is backed by a real data
source; no additional metric should be advertised without implementation
support.
- Around line 625-642: Remove the synthetic sleep fallback block guarded by
xVals.length < 2 in the analysis method, including the generated synthSleep
values and additions to xVals, yVals, and points. When fewer than two real
paired points remain, return the existing insufficient-data error so
correlation, regression, and chart points are never produced from fabricated
health data.
- Around line 714-720: Update get_muscle_group_volume’s matching logic to
resolve each requested group with _resolveMuscleGroup, then compare resolved ids
using _wp.getMuscleGroupName(e.primaryMuscle) instead of raw display-name
substring matching. Preserve valid group requests such as “Quadriceps” and
“Lower Back”, and include secondary muscle activations in the aggregation
alongside each exercise’s primary group.
- Around line 26-28: Update CoachToolService’s constructor to accept the
optional HealthHistoryManager as a named parameter, then update all
instantiations to pass it using healthHistory: while preserving existing
dependency behavior.

In `@workout-logger/lib/services/ai/gemini_ai_service.dart`:
- Around line 91-95: Restrict daily-quota classification in
_isDailyQuotaExhausted within
workout-logger/lib/services/ai/gemini_ai_service.dart:91-95 and
is_daily_quota_exhausted in workout-logger/scripts/test_gemini_api.py:63-64 to
daily-limit identifiers such as GenerateRequestsPerDay or free_tier_requests, or
exact daily metrics; remove generic QuotaExceeded and RESOURCE_EXHAUSTED matches
so minute-scale rate limits continue through retry-delay handling.
- Line 496: Preserve each function-call ID through the response flow: in
workout-logger/lib/services/ai/gemini_ai_service.dart lines 457-496, retain
fc['id'] on each FunctionCall and include the matching ID in every emitted
functionResponse; in workout-logger/scripts/test_gemini_api.py lines 242-255,
copy fc["id"] into each generated functionResponse.
- Around line 98-105: Make the Gemini fallback payload compatible with each
selected model: update _getFallbackModel and the request-building logic around
thinkingConfig so gemini-2.5-flash uses thinkingBudget rather than Gemini 3.x
thinkingLevel, or remove that fallback. Apply the same compatible fallback chain
and per-model thinkingConfig normalization in
workout-logger/lib/services/ai/gemini_ai_service.dart at lines 98-105 and
258-259, and workout-logger/scripts/test_gemini_api.py at lines 67-72 and 221 so
retrying with a new model rebuilds the configuration.

In `@workout-logger/lib/services/interfaces/ml_service_interface.dart`:
- Around line 78-83: Update the documentation for MLService.recommendSets to
state that pastSessions must be ordered most recent first, with index 0 as the
latest prior session and index 1 as the preceding session. Replace the “past 3
sessions” wording with documentation matching the implementation’s two-entry
usage.

In `@workout-logger/lib/services/ml_service.dart`:
- Around line 449-457: Update the post-deload recovery branch in the
recommendation logic to avoid embedding raw set.weight with a hardcoded kg label
in the reasoning string. Either remove the weight value from this message or
route its display through the existing presentation-layer formatting, such as
SettingsProvider.formatWeight, so both units and numeric formatting respect the
user’s settings.
- Around line 375-391: Update deload detection in recommendSets to require the
comparison session to be recent, preventing permanent resets from triggering
recovery from an older workout. Use effective load, consistent with set.volume,
when calculating w0 and w1 for assisted bodyweight exercises instead of raw
set.weight, and cap any recovery recommendations derived from refSets near the
last performed weight rather than allowing high-confidence loads far above
lastSession.

In `@workout-logger/lib/services/settings_provider.dart`:
- Around line 51-53: The body-weight validation in the storage-load path and
setUserBodyWeight must reject non-finite or non-positive values. Update both
paths to accept only finite values greater than zero, persist only valid inputs,
and use 70.0 when the stored userBodyWeight is invalid.

In `@workout-logger/lib/services/workout_provider.dart`:
- Around line 660-671: Update getRecommendations so handle-scoped requests do
not pass the exercise-wide _growthModels[exerciseId] into recommendSets. Key
growth models by both exerciseId and handle when the training contract supports
it; otherwise pass no growth model for handle-scoped recommendations while
preserving the existing exercise-wide behavior for unscoped requests.

In `@workout-logger/lib/theme/a2ui_app_theme.dart`:
- Around line 1-2: Update the imports in the theme adapter to use the public
genui barrel exported by lib/genui/a2ui.dart, importing A2UiTheme through
package:repforge/genui/a2ui.dart instead of the internal src/a2ui_theme.dart
path.

In `@workout-logger/scripts/test_gemini_api.py`:
- Around line 93-98: Update the retry/fallback flow in the test_gemini_api
function so a quota response on the final attempt cannot fall through and return
None after selecting a fallback. Track fallback attempts independently from the
normal retry limit, or raise a clear error when no attempts remain; ensure the
function always returns its expected dict result or an explicit exception before
callers invoke res1.get(...).

In `@workout-logger/test/genui/a2ui_prompt_test.dart`:
- Around line 39-42: Update the example extraction around start and end so the
closing brace search is limited to the region beginning at markerIndex, rather
than using section.lastIndexOf('}') across the entire section. Preserve the
existing substring and validation behavior while ensuring later prompt content
cannot extend the extracted example.

In `@workout-logger/test/genui/a2ui_purity_test.dart`:
- Around line 9-11: Update the _forbiddenPathPattern in the purity test to
include the app-specific data directory alongside theme, models, services, and
screens, ensuring package and relative imports from data are rejected by the
existing guard.
- Around line 95-96: Hoist the cast-matching RegExp used in the loop in the
relevant purity test function into a top-level _castPattern declaration beside
_forbiddenPathPattern and the other shared patterns. Replace the inline RegExp
construction in the per-line check with this shared pattern, preserving the
existing matching expression and behavior.

In `@workout-logger/test/genui/a2ui_registry_test.dart`:
- Around line 133-136: Update the A2UiNode instantiation in the test to use the
const constructor, preserving its existing name and props arguments so
prefer_const_constructors is satisfied.

In `@workout-logger/test/genui/a2ui_renderer_test.dart`:
- Around line 153-158: Update the comment near the GridContainer assertions to
remove the contradictory statement that the node is dropped entirely; document
only that excluding items yields a real zero-children node and the expected
non-crashing behavior, matching the assertions.

In `@workout-logger/test/genui/a2ui_robustness_test.dart`:
- Around line 144-145: Strengthen the minY assertion in the chart axis test to
require the axis minimum to bracket the dataset’s true minimum of -50, rather
than merely being below -10. Keep the existing maxY assertion and the rest of
the chart test unchanged.
- Around line 80-84: Update the test names in both the parser loop and the
“renderer never throws” loop to derive from each payload’s truncated string
instead of the mutable list index. Preserve enough payload content for stable
failure attribution while keeping names concise.

In `@workout-logger/test/genui/a2ui_theme_test.dart`:
- Around line 138-143: Update the Container finders in both assertions around
the panel decoration checks to scope them to the panel widget under test instead
of using the global find.byType(Container). Apply the same scoped finder change
at the second assertion near line 159, preserving the existing padding, color,
and border expectations.

In `@workout-logger/test/genui/components/dynamic_chart_test.dart`:
- Around line 152-161: Add a widget test alongside the existing pie-chart test
that supplies mixed-sign values such as [60, -40], verifies the pie chart
renders without throwing, and confirms the expected fallback behavior from _pie
when percentage calculation receives invalid values.
- Around line 163-175: Extend the testWidgets case for multi-series non-pie
charts to also pump a single-series chart and a multi-series pie chart,
asserting that A2UiLegend is absent in both cases while preserving the existing
positive legend assertions. Add the A2UiLegend import from a2ui_panels.dart and
use it to verify the showLegend condition.

In `@workout-logger/test/genui/components/scatter_plot_test.dart`:
- Around line 89-108: Extend the hostile-input tests around parse to include
points whose x or y coordinates are "NaN" and "Infinity" strings. Assert that
the resulting bounds contain only finite values, ensuring non-finite parsed
doubles are rejected before entering the bounds calculation.

In `@workout-logger/test/genui/components/stat_card_test.dart`:
- Around line 53-56: Add a test case in the “appends a unit that is not already
present” test covering a unit that appears only as a substring, such as value
“10 reps” with unit “s”, and assert the intended formatted result so the
behavior described in stat card parsing is pinned.
- Around line 11-23: Update the test helper pump to pass its props argument into
A2UiProps instead of constructing empty properties, so StatCardSpec.render
receives the caller’s values. Then replace the duplicated pumpWidget setup in
the title/value/subtitle test with a direct pump call using the intended
StatCard properties.

In `@workout-logger/test/new_features_test.dart`:
- Around line 192-231: Extend the CoachToolService test group with cases
covering get_health_metrics, analyze_health_workout_correlation, and
get_muscle_group_volume. Add assertions for the _hh == null error path,
insufficient paired-data behavior, and muscle-group lookup using a display name,
ensuring each test exercises the relevant days window, synthetic-data fallback,
and muscle-id matching defects without changing the existing sleeping-HR test.
- Around line 99-104: Update the assisted pullups test around calculateVolume to
use different values for WorkoutSet.weight and WorkoutSet.assistWeight, while
preserving the expected volume based on bodyweight minus assistWeight plus any
extra weight, multiplied by reps. Choose values that would produce a different
result if weight were incorrectly used in place of assistWeight.

In `@workout-logger/test/settings_provider_test.dart`:
- Line 23: Update SettingsProvider’s _getFallbackModel to use the supported
legacy model gemini-3.5-flash instead of gemini-3.6-flash, while preserving the
existing configuration and AI service behavior.

🪄 Autofix

Fix all unresolved CodeRabbit comments on this PR:

• Push a commit to this branch (recommended)
• Create a new PR with the fixes

---

ℹ️ Review info

⚙️ Run configuration

Configuration used: Organization UI

Review profile: ASSERTIVE

Plan: Pro Plus

Run ID: 74411819-f2bf-4d14-a3f4-c07c5d50e60a

📥 Commits

Reviewing files that changed from the base of the PR and between 39fbb5e and 8024efa.

📒 Files selected for processing (67)

• .github/workflows/release.yml
• .gitignore
• fdroid/metadata/com.devasy.repforge.yml
• scripts/patch_so.py
• workout-logger/lib/data/exercise_database.dart
• workout-logger/lib/genui/a2ui.dart
• workout-logger/lib/genui/src/a2ui_node.dart
• workout-logger/lib/genui/src/a2ui_panels.dart
• workout-logger/lib/genui/src/a2ui_parser.dart
• workout-logger/lib/genui/src/a2ui_prompt.dart
• workout-logger/lib/genui/src/a2ui_props.dart
• workout-logger/lib/genui/src/a2ui_registry.dart
• workout-logger/lib/genui/src/a2ui_renderer.dart
• workout-logger/lib/genui/src/a2ui_series.dart
• workout-logger/lib/genui/src/a2ui_spec.dart
• workout-logger/lib/genui/src/a2ui_theme.dart
• workout-logger/lib/genui/src/components/data_list_group.dart
• workout-logger/lib/genui/src/components/dynamic_chart.dart
• workout-logger/lib/genui/src/components/filter_chips.dart
• workout-logger/lib/genui/src/components/grid_container.dart
• workout-logger/lib/genui/src/components/metric_gauge.dart
• workout-logger/lib/genui/src/components/radar_chart.dart
• workout-logger/lib/genui/src/components/scatter_plot.dart
• workout-logger/lib/genui/src/components/stat_card.dart
• workout-logger/lib/genui/src/default_registry.dart
• workout-logger/lib/main.dart
• workout-logger/lib/models/models.dart
• workout-logger/lib/screens/ai_coach_screen.dart
• workout-logger/lib/screens/widgets/exercise_input_section.dart
• workout-logger/lib/screens/workout_flow_screen.dart
• workout-logger/lib/services/ai/coach_tool_service.dart
• workout-logger/lib/services/ai/gemini_ai_service.dart
• workout-logger/lib/services/gemini_context_builder.dart
• workout-logger/lib/services/interfaces/ai_service_interface.dart
• workout-logger/lib/services/interfaces/ml_service_interface.dart
• workout-logger/lib/services/managers/pr_manager.dart
• workout-logger/lib/services/ml_service.dart
• workout-logger/lib/services/settings_provider.dart
• workout-logger/lib/services/workout_provider.dart
• workout-logger/lib/theme/a2ui_app_theme.dart
• workout-logger/pubspec.yaml
• workout-logger/scripts/test_gemini_api.py
• workout-logger/test/ai_coach_view_model_test.dart
• workout-logger/test/gemini_context_builder_test.dart
• workout-logger/test/genui/a2ui_custom_registry_test.dart
• workout-logger/test/genui/a2ui_parser_test.dart
• workout-logger/test/genui/a2ui_prompt_test.dart
• workout-logger/test/genui/a2ui_props_test.dart
• workout-logger/test/genui/a2ui_purity_test.dart
• workout-logger/test/genui/a2ui_registry_test.dart
• workout-logger/test/genui/a2ui_renderer_test.dart
• workout-logger/test/genui/a2ui_robustness_test.dart
• workout-logger/test/genui/a2ui_series_test.dart
• workout-logger/test/genui/a2ui_theme_test.dart
• workout-logger/test/genui/components/data_list_group_test.dart
• workout-logger/test/genui/components/dynamic_chart_test.dart
• workout-logger/test/genui/components/filter_chips_test.dart
• workout-logger/test/genui/components/metric_gauge_test.dart
• workout-logger/test/genui/components/radar_chart_test.dart
• workout-logger/test/genui/components/scatter_plot_test.dart
• workout-logger/test/genui/components/stat_card_test.dart
• workout-logger/test/new_features_test.dart
• workout-logger/test/routine_optimizer_screen_test.dart
• workout-logger/test/routine_optimizer_view_model_test.dart
• workout-logger/test/screens/ai_coach_genui_test.dart
• workout-logger/test/settings_provider_test.dart
• workout-logger/test/test_utils/mock_ml_service.dart

💤 Files with no reviewable changes (1)

• scripts/patch_so.py

[coderabbitai]

🩺 Stability & Availability | 🟡 Minor | ⚡ Quick win

Guard seriesColor against an empty seriesPalette.

seriesPalette is a required public field with no non-empty constraint. If a host app injects a theme with an empty palette through A2UiThemeProvider, i % seriesPalette.length evaluates i % 0 and throws. Every chart component calls seriesColor during build, so the whole dashboard fails to render. A2UiTheme.dark and repforgeA2UiTheme both supply five colors today, so the fault is not reachable in-repo, but the public constructor allows it.

Add a fallback color, or assert a non-empty palette in the constructor.

🛡️ Proposed guard

   /// Colour for series index [i], cycling through [seriesPalette].
-  Color seriesColor(int i) => seriesPalette[i % seriesPalette.length];
+  Color seriesColor(int i) => seriesPalette.isEmpty
+      ? accent
+      : seriesPalette[i % seriesPalette.length];

📝 Committable suggestion

‼️ IMPORTANT
Carefully review the code before committing. Ensure that it accurately replaces the highlighted code, contains no missing lines, and has no issues with indentation. Thoroughly test & benchmark the code to ensure it meets the requirements.

Suggested change

	/// Colour for series index [i], cycling through [seriesPalette].
	Color seriesColor(int i) => seriesPalette[i % seriesPalette.length];
	/// Colour for series index [i], cycling through [seriesPalette].
	Color seriesColor(int i) => seriesPalette.isEmpty
	? accent
	: seriesPalette[i % seriesPalette.length];

🤖 Prompt for AI Agents

Verify each finding against current code. Fix only still-valid issues, skip the
rest with a brief reason, keep changes minimal, and validate.

In `@workout-logger/lib/genui/src/a2ui_theme.dart` around lines 41 - 42, Guard the
public A2UiTheme palette contract so seriesColor never evaluates a modulo
operation with an empty seriesPalette. Update the A2UiTheme constructor to
assert or otherwise enforce a non-empty palette, while preserving the existing
indexed cycling behavior for valid palettes.

[coderabbitai]

🗄️ Data Integrity & Integration | 🔵 Trivial

🧩 Analysis chain

🏁 Script executed:

#!/bin/bash
# Description: Locate the geminiModel default and every hardcoded model id.
set -euo pipefail

rg -n -C4 'geminiModel' workout-logger/lib --type=dart
rg -n 'gemini-[0-9]' workout-logger --type=dart --type=py --type=yaml

Repository: Devasy/RepForge

Length of output: 7970

---

🌐 Web query:

current available Gemini API model names flash

💡 Result:

As of August 7, 2026, the available Gemini "Flash" series models for the Gemini API include the following primary stable and preview versions: Stable Models gemini-3.6-flash: The latest iteration designed for complex, multi-step workflows, improved code generation, and multimodal reasoning [1][2]. gemini-3.5-flash: Designed for sustained frontier performance in agentic and coding tasks [3][4]. gemini-3.5-flash-lite: Optimized as a low-latency, highly cost-effective option for high-volume automation [3]. gemini-3.1-flash-lite: Optimized for high-volume, cost-sensitive tasks [2][3]. gemini-3.1-flash-lite-image: A multimodal model optimized for low-latency, cost-effective image generation and editing [3]. gemini-2.5-flash: A highly capable model balancing intelligence and latency [2]. gemini-2.5-flash-lite: Built for massive scale and high-throughput tasks [2]. gemini-2.5-flash-image: A native image generation model optimized for speed and contextual understanding [5]. Preview Models gemini-3-flash-preview: Features upgraded visual/spatial reasoning and agentic coding capabilities [3][5]. gemini-omni-flash-preview: Currently available in public preview [3]. gemini-3.1-flash-live-preview: Specifically noted for live interaction capabilities [1]. gemini-3.1-flash-tts-preview: Specifically noted for text-to-speech capabilities [1]. gemini-2.5-flash-live-preview: Specifically noted for live interaction [1]. gemini-2.5-flash-tts-preview: Specifically noted for text-to-speech capabilities [1]. Alias Models The Gemini API also provides convenience aliases that automatically point to the latest release of a model series, such as gemini-flash-latest and gemini-flash-lite-latest [5]. Note: Gemini models are typically accessed in the API using the prefix models/ (e.g., models/gemini-3.6-flash) [6]. Versioning can follow stable, preview, or specific dated patterns [1]. You can verify the most current list of available models programmatically using the models.list method in the Gemini API [6].

Citations:

• 1: https://ai.google.dev/gemini-api/docs/models
• 2: https://docs.cloud.google.com/gemini-enterprise-agent-platform/models/google-models
• 3: https://ai.google.dev/gemini-api/docs/changelog
• 4: https://ai.google.dev/gemini-api/docs/models/gemini-3.5-flash
• 5: https://ai.google.dev/api/interactions-api
• 6: https://ai.google.dev/api/models

---

🌐 Web query:

google_generative_ai 0.4.3 model id gemini-3.6-flash

💡 Result:

The google_generative_ai Dart SDK (version 0.4.3) and the gemini-3.6-flash model are distinct components with different statuses within the Google AI ecosystem as of August 7, 2026. google_generative_ai (SDK) The google_generative_ai package is a legacy Dart SDK that has been deprecated [1][2]. Version 0.4.3 was released approximately two years ago and contained only internal changes to support the Vertex SDK, with no user-visible changes [3]. Because this library is no longer actively maintained and has been deprecated as of November 30, 2025, it does not receive updates to support newer models like Gemini 3.6 Flash [1][2]. Developers are encouraged to migrate to modern alternatives such as Genkit for Dart or Firebase AI Logic [1][2]. gemini-3.6-flash (Model) Gemini 3.6 Flash is a generally available (GA), production-ready model released in July 2026 [4][5][6]. It is part of the Gemini 3 model family and is designed for high efficiency and performance in coding and agentic tasks [7][6]. This model is accessed through the modern Google GenAI SDKs, which replaced the legacy generative AI libraries [8][1]. Using the legacy google_generative_ai SDK to attempt to call gemini-3.6-flash is not the recommended or supported development path [1].

Citations:

• 1: https://ai.google.dev/gemini-api/docs/libraries
• 2: https://pub.dev/documentation/google_generative_ai/latest/
• 3: https://github.com/google-gemini/deprecated-generative-ai-dart/blob/main/pkgs/google_generative_ai/CHANGELOG.md
• 4: https://docs.cloud.google.com/gemini-enterprise-agent-platform/models/gemini/3-6-flash
• 5: https://blog.google/innovation-and-ai/models-and-research/gemini-models/gemini-3-6-flash-3-5-flash-lite-3-5-flash-cyber/
• 6: https://ai.google.dev/gemini-api/docs/changelog
• 7: https://deepmind.google/models/model-cards/gemini-3-6-flash/
• 8: https://ai.google.dev/gemini-api/docs

---

🏁 Script executed:

#!/bin/bash
set -euo pipefail

echo "== pubspec dependency =="
rg -n 'google_generative_ai|google_ai' workout-logger/pubspec.yaml workout-logger/pubspec.lock -C2 || true

echo

echo "== google_generative_ai usages =="
rg -n 'google_generative_ai|Generative|GenerativeModel|googleAI|models\.|generate' workout-logger/lib --type=dart --context=2

Repository: Devasy/RepForge

Length of output: 50371

---

🏁 Script executed:

#!/bin/bash
set -euo pipefail

echo "== AI service model constants/constructor =="
sed -n '1,120p' workout-logger/lib/services/ai/gemini_ai_service.dart

echo

echo "== Settings model conversion around kGeminiModels =="
sed -n '860,920p' workout-logger/lib/screens/widgets/profile_sections.dart

Repository: Devasy/RepForge

Length of output: 7532

---

Fallback before relying on gemini-3.6-flash with this SDK.

SettingsProvider and the AI service now default to gemini-3.6-flash, but google_generative_ai 0.4.3 predates that model. Since the actual call does not construct GenerativeModel and instead posts directly to v1beta/models/gemini-3.6-flash:generateContent, API support may be sufficient, but the local model fallback in _getFallbackModel should use a supported legacy default such as gemini-3.5-flash until the new SDK model is safely available through GenerativeModel.

🤖 Prompt for AI Agents

Verify each finding against current code. Fix only still-valid issues, skip the
rest with a brief reason, keep changes minimal, and validate.

In `@workout-logger/test/settings_provider_test.dart` at line 23, Update
SettingsProvider’s _getFallbackModel to use the supported legacy model
gemini-3.5-flash instead of gemini-3.6-flash, while preserving the existing
configuration and AI service behavior.

[coderabbitai]

Review continued from previous batch...

[coderabbitai]

📐 Maintainability & Code Quality | 🔵 Trivial | 💤 Low value

Hoist the cast pattern out of the loop.

The RegExp is constructed for every line of every component file. Declare it once beside _forbiddenPathPattern, as done for the other two patterns in this file.

♻️ Proposed refactor

Add beside the other patterns:

final RegExp _castPattern =
    RegExp(r'\bas (String|num|int|double|List|Map|bool|Object|dynamic)\b');

Then:

-        if (RegExp(r"\bas (String|num|int|double|List|Map|bool|Object|dynamic)\b")
-            .hasMatch(lines[i])) {
+        if (_castPattern.hasMatch(lines[i])) {
           violations.add('${entity.path}:${i + 1}: ${lines[i].trim()}');
         }

📝 Committable suggestion

‼️ IMPORTANT
Carefully review the code before committing. Ensure that it accurately replaces the highlighted code, contains no missing lines, and has no issues with indentation. Thoroughly test & benchmark the code to ensure it meets the requirements.

Suggested change

	if (RegExp(r"\bas (String|num|int|double|List|Map|bool|Object|dynamic)\b")
	.hasMatch(lines[i])) {
	if (_castPattern.hasMatch(lines[i])) {

🤖 Prompt for AI Agents

Verify each finding against current code. Fix only still-valid issues, skip the
rest with a brief reason, keep changes minimal, and validate.

In `@workout-logger/test/genui/a2ui_purity_test.dart` around lines 95 - 96, Hoist
the cast-matching RegExp used in the loop in the relevant purity test function
into a top-level _castPattern declaration beside _forbiddenPathPattern and the
other shared patterns. Replace the inline RegExp construction in the per-line
check with this shared pattern, preserving the existing matching expression and
behavior.

[coderabbitai]

📐 Maintainability & Code Quality | 🟡 Minor | ⚡ Quick win

The comment contradicts itself and the assertions.

The comment states the node "is dropped entirely rather than silently rendered blank", then states it "actually returns a real zero-children node here". The assertions at lines 159-160 confirm the second statement. Remove the first clause so the comment describes the pinned behavior only.

📝 Proposed fix

-      // GridContainer has no other content, so with `items` correctly
-      // excluded from child resolution it has zero children and is dropped
-      // entirely rather than silently rendered blank — `_declaresChildren`
-      // does not fire for `items`, so this actually returns a real
-      // zero-children node here (GridContainer doesn't declare `items` as
-      // its own data key), which is the expected non-crashing behavior.
+      // `_declaresChildren` does not fire for `items`, so the node is not
+      // rejected: it parses into a real GridContainer with zero children.
+      // That is the expected non-crashing behavior — `items` stays
+      // DataListGroup's own data key and is never read as child components.

📝 Committable suggestion

‼️ IMPORTANT
Carefully review the code before committing. Ensure that it accurately replaces the highlighted code, contains no missing lines, and has no issues with indentation. Thoroughly test & benchmark the code to ensure it meets the requirements.

Suggested change

	// GridContainer has no other content, so with `items` correctly
	// excluded from child resolution it has zero children and is dropped
	// entirely rather than silently rendered blank — `_declaresChildren`
	// does not fire for `items`, so this actually returns a real
	// zero-children node here (GridContainer doesn't declare `items` as
	// its own data key), which is the expected non-crashing behavior.
	// `_declaresChildren` does not fire for `items`, so the node is not
	// rejected: it parses into a real GridContainer with zero children.
	// That is the expected non-crashing behavior — `items` stays
	// DataListGroup's own data key and is never read as child components.

🤖 Prompt for AI Agents

Verify each finding against current code. Fix only still-valid issues, skip the
rest with a brief reason, keep changes minimal, and validate.

In `@workout-logger/test/genui/a2ui_renderer_test.dart` around lines 153 - 158,
Update the comment near the GridContainer assertions to remove the contradictory
statement that the node is dropped entirely; document only that excluding items
yields a real zero-children node and the expected non-crashing behavior,
matching the assertions.

[coderabbitai]

📐 Maintainability & Code Quality | 🔵 Trivial | 💤 Low value

Derive stable test names from the payload.

The names are payload $i. If a payload is inserted in the middle of _payloads, every later test is renamed. That breaks failure attribution across CI runs. Use a truncated payload string in the name instead.

♻️ Proposed refactor

-    for (var i = 0; i < _payloads.length; i++) {
-      test('payload $i', () {
-        expect(() => _parser.parse(_payloads[i]), returnsNormally);
-      });
-    }
+    for (final payload in _payloads) {
+      final label = payload.isEmpty
+          ? '<empty>'
+          : payload.substring(0, payload.length.clamp(0, 60));
+      test('payload $label', () {
+        expect(() => _parser.parse(payload), returnsNormally);
+      });
+    }

Apply the same change to the renderer never throws loop.

🤖 Prompt for AI Agents

Verify each finding against current code. Fix only still-valid issues, skip the
rest with a brief reason, keep changes minimal, and validate.

In `@workout-logger/test/genui/a2ui_robustness_test.dart` around lines 80 - 84,
Update the test names in both the parser loop and the “renderer never throws”
loop to derive from each payload’s truncated string instead of the mutable list
index. Preserve enough payload content for stable failure attribution while
keeping names concise.

[Devasy]

@coderabbitai review

[coderabbitai]

✅ Action performed

Review finished.

Note: CodeRabbit is an incremental review system and does not re-review already reviewed commits. This command is applicable only when automatic reviews are paused.

[coderabbitai]

Actionable comments posted: 10

Caution

Some comments are outside the diff and can’t be posted inline due to platform limitations.

⚠️ Outside diff range comments (6)

workout-logger/lib/services/workout_provider.dart (1)

526-535: 🗄️ Data Integrity & Integration | 🟠 Major | ⚡ Quick win

Preserve the log handle when removing a set.

addSet preserves currentLog.handle, but removeLastSet rebuilds ExerciseLog without handle at Line 548. After a removal, the completed log can lose its handle even when recorded sets remain. Handle-scoped history then cannot match this log.

Proposed fix

         _currentExerciseLogs[_currentExerciseIndex] = ExerciseLog(
           exerciseId: currentLog.exerciseId,
           sets: newSets,
           notes: currentLog.notes,
+          handle: currentLog.handle,
         );

🤖 Prompt for AI Agents

Verify each finding against current code. Fix only still-valid issues, skip the
rest with a brief reason, keep changes minimal, and validate.

In `@workout-logger/lib/services/workout_provider.dart` around lines 526 - 535,
Update removeLastSet to preserve the current ExerciseLog handle when rebuilding
the log, matching the handle retention already implemented in addSet. Pass
currentLog.handle into the replacement ExerciseLog so completed logs with
remaining sets remain associated with handle-scoped history.

workout-logger/lib/screens/widgets/exercise_input_section.dart (1)

129-177: 🎯 Functional Correctness | 🟠 Major | ⚡ Quick win

Label assisted dropsets in _DropsetSection.

When dropsets are enabled for pull-ups, chin-ups, dips, or push-ups, _DropsetSection does not pass isAssistedBodyweightExercise(exerciseId) to its rows, so each weight field is labeled only as the unit, not Assist. Pass the assisted-bodyweight state into _DropsetSection, label each drop weight as Assist (<unit>), and add help text that shows how each assist value maps to effective load.

🤖 Prompt for AI Agents

Verify each finding against current code. Fix only still-valid issues, skip the
rest with a brief reason, keep changes minimal, and validate.

In `@workout-logger/lib/screens/widgets/exercise_input_section.dart` around lines
129 - 177, Update the `_DropsetSection` call and implementation to receive the
assisted-bodyweight state from `isAssistedBodyweightExercise(exerciseId)`. Use
that state when rendering each drop weight field so assisted exercises display
“Assist (<unit>)” instead of only the unit, and add help text explaining the
assist value’s effective-load calculation.

workout-logger/lib/services/ai/coach_tool_service.dart (3)

560-604: 🗄️ Data Integrity & Integration | 🟠 Major | ⚡ Quick win

Use the requested correlation window for sleep data.

The default window is 60 days, but Line 604 always requests HealthGranularity.week. Requests above seven days silently discard older paired days. The raw days value is also not bounded.

Use _limitArg with a supported maximum. Select and trim the sleep-bar granularity as _getHealthMetrics does, or reduce the declared window to seven days. This finding is related to the earlier unbounded health-window finding.

🤖 Prompt for AI Agents

Verify each finding against current code. Fix only still-valid issues, skip the
rest with a brief reason, keep changes minimal, and validate.

In `@workout-logger/lib/services/ai/coach_tool_service.dart` around lines 560 -
604, The correlation flow around the sleepBars call must honor the requested
days window instead of always using HealthGranularity.week. Bound days with
_limitArg using the supported health-window maximum, then select the appropriate
sleep-bar granularity and trim/filter returned sleep data consistently with
_getHealthMetrics before pairing it with workout data.

---

405-524: 🗄️ Data Integrity & Integration | 🟠 Major | ⚡ Quick win

Align the sleeping-HR response with its declared output contract.

The tool declaration promises median and P75 values. This response never returns either value. It also does not return the declared aggregate percentile set.

Return the declared statistics, or remove unsupported fields from the tool description. Otherwise the model can report unavailable analysis as if it were present.

🤖 Prompt for AI Agents

Verify each finding against current code. Fix only still-valid issues, skip the
rest with a brief reason, keep changes minimal, and validate.

In `@workout-logger/lib/services/ai/coach_tool_service.dart` around lines 405 -
524, Update _getSleepingHrAnalytics to match the tool’s declared output contract
by calculating and returning median, P75, and the complete declared aggregate
percentile statistics, including corresponding daily or series values where
required. Reuse the collected sleeping-HR samples and preserve existing fields,
or revise the tool declaration to remove any statistics that cannot be computed;
do not leave declared metrics absent from the response.

---

574-599: 🗄️ Data Integrity & Integration | 🟠 Major | ⚡ Quick win

Aggregate workout values for sessions on the same date.

dayData has one entry per date, but each session assigns m['y'] again. Volume and duration retain only the last session. exercise_max_weight can also replace a higher earlier value with a lower later value.

Sum volume and duration per date. Keep the maximum weight per date.

Proposed fix

-        m['y'] = vol;
+        m['y'] = (m['y'] ?? 0.0) + vol;
...
-        m['y'] = s.duration.toDouble();
+        m['y'] = (m['y'] ?? 0.0) + s.duration;
...
-        if (maxW > 0) m['y'] = maxW;
+        if (maxW > (m['y'] ?? 0.0)) m['y'] = maxW;

🤖 Prompt for AI Agents

Verify each finding against current code. Fix only still-valid issues, skip the
rest with a brief reason, keep changes minimal, and validate.

In `@workout-logger/lib/services/ai/coach_tool_service.dart` around lines 574 -
599, Update the session aggregation loop in the dayData-building logic to
combine values for sessions sharing the same date instead of overwriting m['y'].
Accumulate workout_volume and session_duration into the existing date value,
while exercise_max_weight must retain the maximum of the existing value and the
current session’s max weight. Preserve the existing exercise filtering and omit
zero/unresolved maximum-weight results.

workout-logger/lib/services/ml_service.dart (1)

375-407: 🎯 Functional Correctness | 🟡 Minor | ⚡ Quick win

Use history when lastSession is empty.

If lastSession is empty and pastSessions contains a valid session, this method returns an empty list. refSets starts empty and changes only after a detected deload. Initialize refSets from the first non-empty historical session when no last session exists. Add a regression test for this input.

Proposed fix

-    List<WorkoutSet> refSets = lastSession;
+    List<WorkoutSet> refSets = lastSession.isNotEmpty
+        ? lastSession
+        : pastSessions?.firstWhere(
+              (session) => session.isNotEmpty,
+              orElse: () => const <WorkoutSet>[],
+            ) ??
+            const <WorkoutSet>[];

🤖 Prompt for AI Agents

Verify each finding against current code. Fix only still-valid issues, skip the
rest with a brief reason, keep changes minimal, and validate.

In `@workout-logger/lib/services/ml_service.dart` around lines 375 - 407, Update
the refSets initialization in the surrounding method so that when lastSession is
empty, it falls back to the first non-empty session in pastSessions before the
existing deload detection and empty-result return. Preserve lastSession as the
preferred source when available, and add a regression test covering an empty
lastSession with a valid historical session.

🤖 Prompt for all review comments with AI agents

Verify each finding against current code. Fix only still-valid issues, skip the
rest with a brief reason, keep changes minimal, and validate.

Inline comments:
In
`@docs/superpowers/specs/2026-08-08-sqlite-migration-and-coach-sql-tool-design.md`:
- Around line 189-191: Extend the migration flow in AppInitializer to validate
SQLite data before setting storage_migrated_v1: read back the migrated records
through exportAllData() and compare normalized data, identifiers, counts, nested
sets, JSON blobs, and settings against the Hive source. Set the Hive migration
flag only when validation succeeds; otherwise leave it unset and preserve the
existing migration retry behavior.
- Around line 128-133: Update the personal_records schema so records are keyed
by both exercise_id and handle, and define how existing legacy records without
handles are mapped during migration. Ensure migration preserves separate records
for multiple handles, and add coverage validating that no records collide,
overwrite, or get discarded.
- Around line 186-192: The migration flow in AppInitializer must be atomic and
safely retryable: execute all Hive reads and SQLite writes within one SQLite
transaction, or stage them in a fresh database and replace the active database
only after complete success. Ensure any failure discards all partial writes so
the next launch can rerun from current Hive data without stale rows or
conflicts, and add failure-injection coverage for retries after each entity
boundary.
- Around line 193-194: Update the Hive backup policy in the migration design to
define explicit retention and deletion behavior. Choose either removal after
verified migration with a documented recovery window, or encryption with
inclusion in deletion, reset, and export flows; replace the indefinite-retention
statement and apply the same policy to the related backup lifecycle section.
- Around line 172-180: Add an explicit schema version to
SqliteStorageService.init() by opening the database with version 1 and an
onUpgrade migration callback. Define the migration structure for future schema
changes, and include supported downgrade handling or an explicit data-preserving
down-revision path if sqflite provides downgrade support.
- Around line 204-206: Update the model query execution design to validate
row_limit within 1..500 before applying the outer LIMIT, and add tight bounds
for query length/shape and SQLite execution time to limit unbounded work from
scans, sorts, joins, or recursive CTEs. Preserve the existing error contract by
returning {'error': message} for cap or execution violations.
- Around line 68-72: Update the schema definitions for
exercise_muscle_activations.muscle_group_id, routine_exercises.exercise_id,
sessions.routine_id, exercise_logs.exercise_id, and targets.exercise_id to
include references to their corresponding tables. Also update onConfigure for
every writable connection to execute PRAGMA foreign_keys = ON.
- Around line 203-205: The connection design must not rely on
openReadOnlyDatabase as an independent safety boundary when it can reuse a
writable same-path instance. Update the read-only connection design to open with
singleInstance: false and require an Android integration test confirming writes
fail, or instead use an immutable snapshot/read-only-only handle enforced by
native SQLite.

In `@workout-logger/lib/services/ai/coach_tool_service.dart`:
- Around line 324-330: Remove readiness_score from the metric schema and all
related metric-handling paths in the coach tool service, including the synthetic
calculation near the readiness metric construction. Keep only measured health
metrics such as sleep_hours and deep_sleep_min, and update descriptions or
validation so readiness_score is no longer accepted or presented as independent
health data.
- Around line 1343-1347: Update _limitArg so the clamped value is explicitly
converted to int before returning it, while preserving the existing fallback and
1-to-max bounds.

---

Outside diff comments:
In `@workout-logger/lib/screens/widgets/exercise_input_section.dart`:
- Around line 129-177: Update the `_DropsetSection` call and implementation to
receive the assisted-bodyweight state from
`isAssistedBodyweightExercise(exerciseId)`. Use that state when rendering each
drop weight field so assisted exercises display “Assist (<unit>)” instead of
only the unit, and add help text explaining the assist value’s effective-load
calculation.

In `@workout-logger/lib/services/ai/coach_tool_service.dart`:
- Around line 560-604: The correlation flow around the sleepBars call must honor
the requested days window instead of always using HealthGranularity.week. Bound
days with _limitArg using the supported health-window maximum, then select the
appropriate sleep-bar granularity and trim/filter returned sleep data
consistently with _getHealthMetrics before pairing it with workout data.
- Around line 405-524: Update _getSleepingHrAnalytics to match the tool’s
declared output contract by calculating and returning median, P75, and the
complete declared aggregate percentile statistics, including corresponding daily
or series values where required. Reuse the collected sleeping-HR samples and
preserve existing fields, or revise the tool declaration to remove any
statistics that cannot be computed; do not leave declared metrics absent from
the response.
- Around line 574-599: Update the session aggregation loop in the
dayData-building logic to combine values for sessions sharing the same date
instead of overwriting m['y']. Accumulate workout_volume and session_duration
into the existing date value, while exercise_max_weight must retain the maximum
of the existing value and the current session’s max weight. Preserve the
existing exercise filtering and omit zero/unresolved maximum-weight results.

In `@workout-logger/lib/services/ml_service.dart`:
- Around line 375-407: Update the refSets initialization in the surrounding
method so that when lastSession is empty, it falls back to the first non-empty
session in pastSessions before the existing deload detection and empty-result
return. Preserve lastSession as the preferred source when available, and add a
regression test covering an empty lastSession with a valid historical session.

In `@workout-logger/lib/services/workout_provider.dart`:
- Around line 526-535: Update removeLastSet to preserve the current ExerciseLog
handle when rebuilding the log, matching the handle retention already
implemented in addSet. Pass currentLog.handle into the replacement ExerciseLog
so completed logs with remaining sets remain associated with handle-scoped
history.

🪄 Autofix

Fix all unresolved CodeRabbit comments on this PR:

• Push a commit to this branch (recommended)
• Create a new PR with the fixes

---

ℹ️ Review info

⚙️ Run configuration

Configuration used: Organization UI

Review profile: ASSERTIVE

Plan: Pro Plus

Run ID: 41c03bb5-3016-4f57-a26a-90de8fe188b3

📥 Commits

Reviewing files that changed from the base of the PR and between 8024efa and 0ff6c2d.

📒 Files selected for processing (29)

• .github/workflows/release.yml
• docs/superpowers/specs/2026-08-08-sqlite-migration-and-coach-sql-tool-design.md
• workout-logger/lib/genui/src/a2ui_panels.dart
• workout-logger/lib/genui/src/a2ui_parser.dart
• workout-logger/lib/genui/src/a2ui_renderer.dart
• workout-logger/lib/genui/src/a2ui_spec.dart
• workout-logger/lib/genui/src/a2ui_theme.dart
• workout-logger/lib/genui/src/components/dynamic_chart.dart
• workout-logger/lib/genui/src/components/metric_gauge.dart
• workout-logger/lib/genui/src/components/stat_card.dart
• workout-logger/lib/main.dart
• workout-logger/lib/models/models.dart
• workout-logger/lib/screens/widgets/exercise_input_section.dart
• workout-logger/lib/screens/workout_flow_screen.dart
• workout-logger/lib/services/ai/coach_tool_service.dart
• workout-logger/lib/services/ai/gemini_ai_service.dart
• workout-logger/lib/services/interfaces/ml_service_interface.dart
• workout-logger/lib/services/ml_service.dart
• workout-logger/lib/services/settings_provider.dart
• workout-logger/lib/services/workout_provider.dart
• workout-logger/lib/theme/a2ui_app_theme.dart
• workout-logger/scripts/test_gemini_api.py
• workout-logger/test/genui/a2ui_prompt_test.dart
• workout-logger/test/genui/a2ui_purity_test.dart
• workout-logger/test/genui/a2ui_robustness_test.dart
• workout-logger/test/genui/a2ui_theme_test.dart
• workout-logger/test/genui/components/dynamic_chart_test.dart
• workout-logger/test/genui/components/stat_card_test.dart
• workout-logger/test/new_features_test.dart

[coderabbitai]

🗄️ Data Integrity & Integration | 🟠 Major | ⚡ Quick win

🧩 Analysis chain

🏁 Script executed:

#!/bin/bash
set -euo pipefail

echo "Files matching SQLite migration/coach SQL spec:"
fd -i '2026-08-08-sqlite-migration-and-coach-sql-tool-design.md|sqflite|sqlite' . | sed -n '1,120p'

echo
echo "Target file excerpt:"
if [ -f "docs/superpowers/specs/2026-08-08-sqlite-migration-and-coach-sql-tool-design.md" ]; then
  nl -ba docs/superpowers/specs/2026-08-08-sqlite-migration-and-coach-sql-tool-design.md | sed -n '55,130p'
fi

echo
echo "Search for foreign key declarations/pragmas/onConfigure:"
rg -n "FOREIGN KEY|PRAGMA foreign_keys|foreign_keys|onConfigure|REFERENCES\s*\(" -S . \
  --glob '!**/.git/**' \
  --glob '!**/build/**' \
  --glob '!**/dist/**' \
  | sed -n '1,240p'

Repository: Devasy/RepForge

Length of output: 346

---

🏁 Script executed:

#!/bin/bash
set -euo pipefail

echo "Target file excerpt:"
if [ -f "docs/superpowers/specs/2026-08-08-sqlite-migration-and-coach-sql-tool-design.md" ]; then
  awk 'NR>=55 && NR<=130 {printf "%6d\t%s\n", NR, $0}' docs/superpowers/specs/2026-08-08-sqlite-migration-and-coach-sql-tool-design.md
fi

echo
echo "Search for foreign key declarations/pragmas/onConfigure:"
rg -n --fixed-strings "FOREIGN KEY" "PRAGMA foreign_keys" "foreign_keys" "onConfigure" "REFERENCES (" "REFERENCES(" . \
  --glob '!**/.git/**' \
  --glob '!**/build/**' \
  --glob '!**/dist/**' \
  || true

echo
echo "Candidate database files:"
fd -i 'database|db|sqlite|sqflite' . \
  --glob '!**/.git/**' \
  --glob '!**/build/**' \
  --glob '!**/dist/**' \
  | sed -n '1,80p'

Repository: Devasy/RepForge

Length of output: 3203

---

🌐 Web query:

SQLite foreign key constraints enabled per database connection PRAGMA foreign_keys ON documentation

💡 Result:

In SQLite, foreign key constraints must be enabled for each database connection using the PRAGMA foreign_keys = ON; command [1][2]. Key details regarding foreign key enforcement include: 1. Default Setting: Foreign key enforcement is disabled by default for backward compatibility [1][2]. Applications should explicitly enable or disable this setting as needed rather than relying on the default, as future releases may change this behavior [3][1]. 2. Connection-Specific: The setting applies only to the current database connection [1][4]. 3. Constraints on Use: The PRAGMA foreign_keys command is a no-op if executed within a transaction (i.e., when a BEGIN or SAVEPOINT is pending) [3][5]. 4. Compilation Requirements: For foreign key constraints to function, the SQLite library must be compiled without the SQLITE_OMIT_FOREIGN_KEY or SQLITE_OMIT_TRIGGER preprocessor macros [1][2]. 5. Alternative Configuration: Developers can also use the sqlite3_db_config C interface with the SQLITE_DBCONFIG_ENABLE_FKEY option to enable or disable foreign key enforcement programmatically [6]. To verify the current status of foreign key enforcement, you can execute the command PRAGMA foreign_keys;, which returns 1 if enabled or 0 if disabled [1][2]. If the command returns no data, it indicates that the version of SQLite being used does not support foreign key enforcement [1][2].

Citations:

• 1: http://sqlite.org/foreignkeys.html
• 2: https://sqlite.org/draft/foreignkeys.html
• 3: https://sqlite.org/pragma.html
• 4: https://www2.sqlite.org/matrix/foreignkeys.html
• 5: https://sqlite.org/draft/pragma.html
• 6: https://www.sqlite.org/draft/c3ref/c_dbconfig_enable_fkey.html

---

Add the missing foreign keys and enable enforcement.

Add foreign keys for muscle_group_id, routine_exercises.exercise_id, sessions.routine_id, exercise_logs.exercise_id, and targets.exercise_id.

Also execute PRAGMA foreign_keys = ON in onConfigure for every writable connection. SQLite disables foreign-key enforcement by default and requires per-connection enablement.

🤖 Prompt for AI Agents

Verify each finding against current code. Fix only still-valid issues, skip the
rest with a brief reason, keep changes minimal, and validate.

In
`@docs/superpowers/specs/2026-08-08-sqlite-migration-and-coach-sql-tool-design.md`
around lines 68 - 72, Update the schema definitions for
exercise_muscle_activations.muscle_group_id, routine_exercises.exercise_id,
sessions.routine_id, exercise_logs.exercise_id, and targets.exercise_id to
include references to their corresponding tables. Also update onConfigure for
every writable connection to execute PRAGMA foreign_keys = ON.

Source: MCP tools

[coderabbitai]

🗄️ Data Integrity & Integration | 🟠 Major | 🏗️ Heavy lift

Preserve the handle dimension in personal records.

personal_records.exercise_id permits only one record per exercise. The current handle-aware workout flow requires separate records for different handles. Migration will therefore collide, overwrite, or discard data when one exercise has multiple handles.

Add handle to the key or unique constraint. Define the legacy-record mapping and test migration with multiple handles.

🤖 Prompt for AI Agents

Verify each finding against current code. Fix only still-valid issues, skip the
rest with a brief reason, keep changes minimal, and validate.

In
`@docs/superpowers/specs/2026-08-08-sqlite-migration-and-coach-sql-tool-design.md`
around lines 128 - 133, Update the personal_records schema so records are keyed
by both exercise_id and handle, and define how existing legacy records without
handles are mapped during migration. Ensure migration preserves separate records
for multiple handles, and add coverage validating that no records collide,
overwrite, or get discarded.

[coderabbitai]

🗄️ Data Integrity & Integration | 🟠 Major | ⚡ Quick win

🧩 Analysis chain

🏁 Script executed:

#!/bin/bash
set -euo pipefail

# Locate the reviewed file and nearby sqlite storage related files.
fd -a '2026-08-08-sqlite-migration-and-coach-sql-tool-design.md|sqlite_storage_service|IStorageService|storage_service' . | sed 's#^\./##' | sort

echo '--- reviewed file excerpts ---'
file='docs/superpowers/specs/2026-08-08-sqlite-migration-and-coach-sql-tool-design.md'
if [ -f "$file" ]; then
  sed -n '130,210p' "$file"
fi

echo '--- git diff stat ---'
git diff --stat || true

echo '--- search openDatabase versions in tracked files ---'
rg -n "openDatabase|onCreate|onUpgrade|version:\s*[0-9]+|version:" --glob '*.dart' . || true

Repository: Devasy/RepForge

Length of output: 6778

---

Add an explicit SQLite schema version and upgrade path.

SqliteStorageService.init() only describes onCreate; later app versions must also get schema changes. Add version: 1 and define onUpgrade migrations, including downgrade behavior if sqflite supports it or an explicit data-preserving down-revision path.

🤖 Prompt for AI Agents

Verify each finding against current code. Fix only still-valid issues, skip the
rest with a brief reason, keep changes minimal, and validate.

In
`@docs/superpowers/specs/2026-08-08-sqlite-migration-and-coach-sql-tool-design.md`
around lines 172 - 180, Add an explicit schema version to
SqliteStorageService.init() by opening the database with version 1 and an
onUpgrade migration callback. Define the migration structure for future schema
changes, and include supported downgrade handling or an explicit data-preserving
down-revision path if sqflite provides downgrade support.

Source: MCP tools

[coderabbitai]

🗄️ Data Integrity & Integration | 🟠 Major | 🏗️ Heavy lift

Make migration retries atomic and idempotent.

A failure after some entity writes leaves partial SQLite state. The next attempt then writes against that state. If Hive changed during the fallback launch, stale rows and primary-key conflicts can remain.

Wrap the complete migration in one SQLite transaction, or migrate into a fresh staging database and replace it only after success. Add a failure-injection test that retries after each entity boundary.

Also applies to: 214-214

🤖 Prompt for AI Agents

Verify each finding against current code. Fix only still-valid issues, skip the
rest with a brief reason, keep changes minimal, and validate.

In
`@docs/superpowers/specs/2026-08-08-sqlite-migration-and-coach-sql-tool-design.md`
around lines 186 - 192, The migration flow in AppInitializer must be atomic and
safely retryable: execute all Hive reads and SQLite writes within one SQLite
transaction, or stage them in a fresh database and replace the active database
only after complete success. Ensure any failure discards all partial writes so
the next launch can rerun from current Hive data without stale rows or
conflicts, and add failure-injection coverage for retries after each entity
boundary.

[coderabbitai]

🗄️ Data Integrity & Integration | 🟠 Major | 🏗️ Heavy lift

Validate migrated data before setting storage_migrated_v1.

A migration that completes without throwing can still lose fields during model-to-table conversion. After the flag is set, SQLite becomes authoritative.

Read back the migrated data and compare a normalized exportAllData() result, entity identifiers, counts, nested sets, JSON blobs, and settings. Set the Hive flag only after validation succeeds.

🤖 Prompt for AI Agents

Verify each finding against current code. Fix only still-valid issues, skip the
rest with a brief reason, keep changes minimal, and validate.

In
`@docs/superpowers/specs/2026-08-08-sqlite-migration-and-coach-sql-tool-design.md`
around lines 189 - 191, Extend the migration flow in AppInitializer to validate
SQLite data before setting storage_migrated_v1: read back the migrated records
through exportAllData() and compare normalized data, identifiers, counts, nested
sets, JSON blobs, and settings against the Hive source. Set the Hive migration
flag only when validation succeeds; otherwise leave it unset and preserve the
existing migration retry behavior.

[coderabbitai]

🔒 Security & Privacy | 🟠 Major | 🏗️ Heavy lift

Define retention and deletion behavior for the Hive backup.

Keeping Hive indefinitely retains stale sessions, conversations, and settings after SQLite deletes or changes them. A user reset or data-deletion flow can therefore leave sensitive data on disk.

Either remove Hive after verified migration with a documented recovery window, or encrypt it and include it in deletion, reset, and export lifecycle handling.

Also applies to: 220-224

🤖 Prompt for AI Agents

Verify each finding against current code. Fix only still-valid issues, skip the
rest with a brief reason, keep changes minimal, and validate.

In
`@docs/superpowers/specs/2026-08-08-sqlite-migration-and-coach-sql-tool-design.md`
around lines 193 - 194, Update the Hive backup policy in the migration design to
define explicit retention and deletion behavior. Choose either removal after
verified migration with a documented recovery window, or encryption with
inclusion in deletion, reset, and export flows; replace the indefinite-retention
statement and apply the same policy to the related backup lifecycle section.

[coderabbitai]

🔒 Security & Privacy | 🟠 Major | ⚡ Quick win

🧩 Analysis chain

🏁 Script executed:

#!/usr/bin/env bash
set -euo pipefail
rg -n 'openReadOnlyDatabase|singleInstance|readOnly' workout-logger/lib workout-logger/test

Repository: Devasy/RepForge

Length of output: 153

---

🏁 Script executed:

#!/usr/bin/env bash
set -euo pipefail

echo "== repo files matching sqlite/spec docs =="
git ls-files | rg '(^|/)(specs|.*sqlite.*|sql.*|storage|.*storage.*|.*Sql|.*sql.*)' | head -200

echo
echo "== locate target file =="
fd -a '2026-08-08-sqlite-migration-and-coach-sql-tool-design\.md|sqflite|SqliteStorageService|storage_service|storage' . | head -200

echo
echo "== target context if present =="
if [ -f "docs/superpowers/specs/2026-08-08-sqlite-migration-and-coach-sql-tool-design.md" ]; then
  nl -ba docs/superpowers/specs/2026-08-08-sqlite-migration-and-coach-sql-tool-design.md | sed -n '170,230p'
fi

Repository: Devasy/RepForge

Length of output: 1270

---

🏁 Script executed:

#!/usr/bin/env bash
set -euo pipefail

echo "== search any openReadOnlyDatabase in repo (broad) =="
rg -n 'openReadOnlyDatabase|openDatabase.*readOnly|readOnlyDatabase|singleInstance|readOnly' . --glob '!**/.git/**' || true

echo
echo "== search SqliteStorageService =="
rg -n 'SqliteStorageService|sqliteStorage|sqlite' workout-logger/lib workout-logger/test workout-logger/flutter_runner workout-logger/ios workout-logger/android 2>/dev/null || true

Repository: Devasy/RepForge

Length of output: 615

---

🏁 Script executed:

#!/usr/bin/env bash
set -euo pipefail

python3 - <<'PY'
from pathlib import Path
p = Path('docs/superpowers/specs/2026-08-08-sqlite-migration-and-coach-sql-tool-design.md')
lines = p.read_text(encoding='utf-8').splitlines()
for i in range(202, 217):
    print(f"{i+1}: {lines[i]}")
PY

echo
echo "== current sqflite declarations in pubspecs =="
git ls-files '*pubspec.yaml' | xargs -r sh -c 'echo "--- $0"; grep -n -i "sqflite\|sqlite" "$0" || true'

Repository: Devasy/RepForge

Length of output: 2525

---

🌐 Web query:

sqflite openReadOnlyDatabase singleInstance readOnly openDatabase documentation

💡 Result:

In the sqflite package, there are two primary ways to open a database in read-only mode: using the dedicated openReadOnlyDatabase function or the standard openDatabase function with the readOnly parameter [1][2][3]. openReadOnlyDatabase The openReadOnlyDatabase function is a convenience method specifically designed to open a database at a given path in read-only mode [1][4]. Its implementation internally calls openDatabase with readOnly set to true [1]. Signature: Future openReadOnlyDatabase(String path, {bool? singleInstance = true}) openDatabase Alternatively, you can use the standard openDatabase function by setting the readOnly parameter to true [3][5]. Signature: Future openDatabase(String path, {bool? readOnly = false, bool? singleInstance = true,...}) When readOnly is set to true, the database is opened as-is, and all other configuration parameters (such as version, onCreate, or onUpgrade callbacks) are ignored [3][5]. The singleInstance Parameter The singleInstance parameter, which defaults to true, determines how the database instance is managed [3][5]. When singleInstance is true, sqflite returns a shared database instance for a given path [3][5]. If you call openDatabase or openReadOnlyDatabase multiple times with the same path, the library returns this same existing instance [3][5]. If singleInstance is set to false, a new connection/instance is created each time, which can be useful in specific scenarios like multi-isolate access, though developers are generally advised to manage their own global reference to avoid lock issues [6]. Summary of Behavior When you set readOnly: true (or use the helper function), you cannot perform write operations, and database initialization callbacks like onCreate or onUpgrade will not be executed [3][7][5]. This is commonly used for checking if a file is a valid database file by attempting to open it and read its version [8].

Citations:

• 1: https://pub.dev/documentation/sqflite/latest/sqflite/openReadOnlyDatabase.html
• 2: https://pub.dev/documentation/sqflite/latest/sqflite/
• 3: https://pub.dev/documentation/sqflite/latest/sqflite/openDatabase.html
• 4: https://pub.dev/documentation/sqflite_common/latest/sqflite/openReadOnlyDatabase.html
• 5: https://pub.dev/documentation/sqflite_common/latest/sqflite/openDatabase.html
• 6: https://github.com/tekartik/sqflite/blob/master/sqflite/doc/usage_recommendations.md
• 7: https://pub.dev/documentation/sqflite/latest/sqflite/OpenDatabaseOptions-class.html
• 8: https://github.com/tekartik/sqflite/blob/master/sqflite/doc/opening_db.md

---

🏁 Script executed:

#!/usr/bin/env bash
set -euo pipefail

python3 - <<'PY'
from pathlib import Path
p = Path('docs/superpowers/specs/2026-08-08-sqlite-migration-and-coach-sql-tool-design.md')
text = p.read_text(encoding='utf-8')
checks = {
    'contains_openReadOnlyDatabase': 'openReadOnlyDatabase' in text,
    'claims_primary_safety_boundary': 'real safety boundary' in text,
    'mentions_single_instance_arg': 'singleInstance' in text,
    'uses_same_file': 'same database file' in text,
}
for name, ok in checks.items():
    print(f"{name}: {ok}")
PY

echo
echo "== inspect current sqflite declaration in pubspec =="
python3 - <<'PY'
from pathlib import Path
p = Path('workout-logger/pubspec.yaml')
print(p.read_text(encoding='utf-8'))
PY

Repository: Devasy/RepForge

Length of output: 4424

---

Do not present openReadOnlyDatabase on the same file as an independent safety boundary.

Because openReadOnlyDatabase forwards singleInstance: true, a later same-path open can return the existing writable database instance, so the SQL tool’s readonly option can be discarded. Open with singleInstance: false and require an Android integration test that write operations fail, or use an immutable snapshot/read-only-only handle that is enforced by the native SQLite layer.

🤖 Prompt for AI Agents

Verify each finding against current code. Fix only still-valid issues, skip the
rest with a brief reason, keep changes minimal, and validate.

In
`@docs/superpowers/specs/2026-08-08-sqlite-migration-and-coach-sql-tool-design.md`
around lines 203 - 205, The connection design must not rely on
openReadOnlyDatabase as an independent safety boundary when it can reuse a
writable same-path instance. Update the read-only connection design to open with
singleInstance: false and require an Android integration test confirming writes
fail, or instead use an immutable snapshot/read-only-only handle enforced by
native SQLite.

Source: MCP tools

[coderabbitai]

🚀 Performance & Scalability | 🟠 Major | ⚡ Quick win

🧩 Analysis chain

🏁 Script executed:

#!/bin/bash
set -euo pipefail

echo "== file excerpt =="
sed -n '160,240p' docs/superpowers/specs/2026-08-08-sqlite-migration-and-coach-sql-tool-design.md || true

echo
echo "== related mentions =="
rg -n "Row cap|LIMIT|validate|timeout|query length|500|200" docs/superpowers/specs/2026-08-08-sqlite-migration-and-coach-sql-tool-design.md . --glob '!.git/**' | head -120

echo
echo "== local sqlite behavior probe: outer LIMIT does not limit work and negative LIMIT has no upper bound =="
python3 - <<'PY'
import sqlite3
stmt = [
    """CREATE TABLE IF NOT EXISTS t(i INTEGER PRIMARY KEY, v TEXT);""",
    """INSERT INTO t(i, v) SELECT i, printf('%d', i) FROM (WITH RECURSIVE c(x) AS (VALUES(1) UNION ALL SELECT x+1 FROM c WHERE x < :n) SELECT x FROM c)""",
    """SELECT * FROM (SELECT v FROM t ORDER BY v, i DESC) LIMIT ?;""",
]
conn = sqlite3.connect(":memory:")
conn.execute("""CREATE TABLE t(i INTEGER PRIMARY KEY, v TEXT);""")
n = 8000
queries = []
for i, q in enumerate(stmt, start=1):
    s = q[:2000]
    print(f"q{i}: {s[:160].replace(chr(10), ' ')}...")
PY

Repository: Devasy/RepForge

Length of output: 20188

---

🌐 Web query:

SQLite LIMIT clause negative value -1 no limit behavior documentation

💡 Result:

In SQLite, providing a negative value to the LIMIT clause is interpreted as "no limit," meaning there is no upper bound on the number of rows returned or affected [1][2][3]. This behavior applies consistently across SELECT statements [1][2], as well as UPDATE and DELETE statements (when those statements are used in configurations that support LIMIT clauses, such as when SQLite is compiled with SQLITE_ENABLE_UPDATE_DELETE_LIMIT) [3]. If a negative value is provided, the query proceeds as if the LIMIT clause were absent, returning or processing the entire result set [2][3].

Citations:

• 1: https://system.data.sqlite.org/src/doc/dc206da59f/Doc/Extra/lang_select.html
• 2: https://stackoverflow.com/questions/6329032/sqlite3-and-limiting-the-number-of-results
• 3: https://system.data.sqlite.org/home/doc/0a3d6229a7425242/Doc/Extra/Core/lang_update.html

---

Bound accepted row caps and SQL execution budgets.

The outer LIMIT only limits rows returned, not the work SQLite does for scans, sorts, cross joins, or recursive CTEs. SQLite also treats negative LIMIT values as no upper bound. Require row_limit in 1..500 before wrapping the query, keep query length/shape/execution time tight, or snapshot the database for read-heavy tool output.

🤖 Prompt for AI Agents

Verify each finding against current code. Fix only still-valid issues, skip the
rest with a brief reason, keep changes minimal, and validate.

In
`@docs/superpowers/specs/2026-08-08-sqlite-migration-and-coach-sql-tool-design.md`
around lines 204 - 206, Update the model query execution design to validate
row_limit within 1..500 before applying the outer LIMIT, and add tight bounds
for query length/shape and SQLite execution time to limit unbounded work from
scans, sorts, joins, or recursive CTEs. Preserve the existing error contract by
returning {'error': message} for cap or execution violations.

Source: MCP tools

[coderabbitai]

🗄️ Data Integrity & Integration | 🟠 Major | ⚡ Quick win

Remove the synthetic readiness_score metric.

At Line 613, the implementation creates readiness_score from sleep duration with an arbitrary 70–100 formula. This is not a measured readiness value. A correlation result can therefore present a sleep-derived proxy as independent health data.

Remove readiness_score from the schema, or rename it as an explicitly documented sleep-derived proxy. This finding is related to the earlier health-metric contract finding.

🤖 Prompt for AI Agents

Verify each finding against current code. Fix only still-valid issues, skip the
rest with a brief reason, keep changes minimal, and validate.

In `@workout-logger/lib/services/ai/coach_tool_service.dart` around lines 324 -
330, Remove readiness_score from the metric schema and all related
metric-handling paths in the coach tool service, including the synthetic
calculation near the readiness metric construction. Keep only measured health
metrics such as sleep_hours and deep_sleep_min, and update descriptions or
validation so readiness_score is no longer accepted or presented as independent
health data.

[coderabbitai]

🎯 Functional Correctness | 🔴 Critical | ⚡ Quick win

🧩 Analysis chain

🏁 Script executed:

#!/bin/bash
set -euo pipefail

rg -n -C3 'int _limitArg|return n\.clamp' \
  workout-logger/lib/services/ai/coach_tool_service.dart

Repository: Devasy/RepForge

Length of output: 647

---

🌐 Web query:

Dart num.clamp int static return type documentation

💡 Result:

The Dart num.clamp method is defined in the dart:core library with the static return type num [1][2][3]. While its formal signature in the num class is num clamp(num lowerLimit, num upperLimit), the method is designed to return the value of the receiver or the specified limits, ensuring the result stays within the inclusive range [lowerLimit, upperLimit] [2][3]. It is important to note the following regarding its return type and behavior: 1. Static Type: The declared return type is num [1][3]. Consequently, when using this method, the static analyzer will treat the result as a num, even if the runtime value is specifically an int or a double [4]. 2. Runtime Behavior: Because num is a common supertype for both int and double, the actual runtime type of the returned object will depend on the types of the receiver, the lowerLimit, and the upperLimit [4]. If all inputs are integers, the result will typically be an int; if any input is a double, the result will typically be a double [4]. 3. Special Handling: There have been discussions within the Dart community regarding the limitations of this signature—specifically that it does not provide more precise static return type information (e.g., returning int when only integers are involved) [4]. However, the signature remains num for backward compatibility and to maintain the flexibility of the num type [4]. When working with clamp, you may need to perform explicit type casting or use methods like toInt if you require a specific numeric type for subsequent operations [2][3].

Citations:

• 1: https://api.dart.dev/dev/2.0.0-dev.67.0/dart-core/num-class.html
• 2: https://github.com/dart-lang/sdk/blob/796d6cc344b93c0432b404173438e0343fc6c2b8/sdk/lib/core/num.dart
• 3: https://github.com/dart-lang/sdk/blob/c42a304ce78f37b717cc055abc503023871a7013/sdk/lib/core/num.dart
• 4: double.clamp should return a double dart-lang/sdk#39652

---

Return an int from _limitArg.

n.clamp(1, max) has static type num, but _limitArg returns int, so static analysis reports a return-type mismatch.

Proposed fix

-    return n.clamp(1, max);
+    if (n < 1) return 1;
+    if (n > max) return max;
+    return n;

📝 Committable suggestion

‼️ IMPORTANT
Carefully review the code before committing. Ensure that it accurately replaces the highlighted code, contains no missing lines, and has no issues with indentation. Thoroughly test & benchmark the code to ensure it meets the requirements.

Suggested change

	int _limitArg(Map<String, Object?> args, int fallback,
	{String key = 'limit', int max = 40}) {
	final n = (args[key] as num?)?.toInt();
	if (n == null) return fallback;
	return n.clamp(1, 40);
	return n.clamp(1, max);
	int _limitArg(Map<String, Object?> args, int fallback,
	{String key = 'limit', int max = 40}) {
	final n = (args[key] as num?)?.toInt();
	if (n == null) return fallback;
	if (n < 1) return 1;
	if (n > max) return max;
	return n;

🤖 Prompt for AI Agents

Verify each finding against current code. Fix only still-valid issues, skip the
rest with a brief reason, keep changes minimal, and validate.

In `@workout-logger/lib/services/ai/coach_tool_service.dart` around lines 1343 -
1347, Update _limitArg so the clamped value is explicitly converted to int
before returning it, while preserving the existing fallback and 1-to-max bounds.