Skip to content

Site: rename to PolicyEngine Macro + P0 accuracy/honesty pass#5

Open
vahid-ahmadi wants to merge 1 commit into
mainfrom
policyengine-macro-site
Open

Site: rename to PolicyEngine Macro + P0 accuracy/honesty pass#5
vahid-ahmadi wants to merge 1 commit into
mainfrom
policyengine-macro-site

Conversation

@vahid-ahmadi

Copy link
Copy Markdown
Contributor

Summary

Implements the P0 tier of the improvement plan (from the Slack thread): rename to PolicyEngine Macro and fix inaccurate positioning/claims across all seven site pages. Every change is grounded in the models' actual verified behaviour, in the same honesty spirit as the OBR/BoE CI work.

A Vercel preview deployment will build for this PR — please review the rendered pages there.

What changed

Rename (product name only — the macromod CLI/package, MCP server name, and repo URL are unchanged): macromod/MacroModPolicyEngine Macro in titles, meta, nav/footer brand, and prose.

Accuracy / honesty (P0):

  • Homepage — new hero ("Open models for the UK economy"); positioned as PolicyEngine's macro suite (not "one reform scored four ways"). Removed the false "every model scores the same PolicyEngine reform objects." Replaced "Statute in, economy out." with a broader analysis workflow.
  • Status badges replace the blanket "live": OBR/SVAR = hosted + validated replication; OG-UK = local only + experimental; PolicyEngine = hosted: household + population planned.
  • OBR page — dropped "the actual one" / "the OBR's own model" → "an open Python implementation of the OBR's published equations"; added an independent-implementation disclaimer and an honest anchored==EFO / raw-calibration status note.
  • SVAR page — "right now" → latest data vintage (2024 Q2); data-vintage note; independent-replication disclaimer (Brignone & Piffer 2025; not endorsed by the Bank of England).
  • OG-UK page — local-only + experimental; honest runtimes (17+ min steady state, hours for transitions); verification grade (structural counterfactual, no ground truth); broadened structural capabilities.
  • PolicyEngine page — household (hosted) vs population (planned/local) made explicit.
  • docs — reframed opening; "right now" → vintage; added the verification-gradient statement (replications vs structural counterfactual).
  • use-it page — confirmed illustrative outputs are labelled; clarified hosted vs local-only tools.
  • style.css — new access/status badge classes.

Out of scope (left for review)

P1–P3 of the plan — Questions page, capability registry, deterministic question router, common request/result schema, and the model-comparison framework — are not in this PR. This is the "correct and clarify" sprint; the bigger product architecture should be reviewed before building.

🤖 Generated with Claude Code

Implements the P0 tier of Max's improvement plan — correct positioning and
remove inaccurate claims — across all seven pages. Grounded in the models'
actual verified behaviour.

Rename (product only; CLI `macromod`, package, repo URL, and MCP server name
unchanged): macromod/MacroMod -> PolicyEngine Macro in titles, meta, nav/footer
brand, and prose.

Accuracy / honesty fixes:
- Homepage: new hero ("Open models for the UK economy"), broader positioning
  (macro suite under the PolicyEngine umbrella; not one reform scored four
  ways). Removed the false "every model scores the same PolicyEngine reform
  objects." Replaced "Statute in, economy out." with "Different questions need
  different models" and a broader analysis workflow.
- Accurate model badges replacing the blanket "live": OBR/SVAR = hosted +
  validated replication; OG-UK = local only + experimental; PolicyEngine =
  hosted (household) + population planned.
- OBR page: dropped "the actual one" / "the OBR's own model" overclaims ->
  "an open Python implementation of the OBR's published equations", added an
  independent-implementation disclaimer (not produced/endorsed by the OBR) and
  an honest anchored==EFO / raw-calibration status note.
- SVAR page: "right now" -> latest data vintage (packaged data end 2024 Q2),
  data-vintage note, and an independent-replication disclaimer (Brignone &
  Piffer 2025; not endorsed by the Bank of England).
- OG-UK page: local-only + experimental status, honest runtimes (17+ min
  steady state, hours for transitions), verification grade (structural
  counterfactual, no ground truth), broadened structural capabilities.
- PolicyEngine page: household (hosted) vs population (planned/local) made
  explicit.
- docs: reframed opening away from "scores a policy reform", "right now" ->
  vintage, added the verification-gradient statement (replications vs
  structural counterfactual).
- Use-it page: confirmed illustrative outputs are labelled; clarified which
  tools are hosted vs local-only.
- style.css: new access/status badge classes (hosted/local/planned/
  experimental/replication).

P1-P3 of the plan (Questions page, capability registry, question router,
common result schema, comparison framework) are intentionally out of scope
here and left for review.

Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
@vercel

vercel Bot commented Jul 15, 2026

Copy link
Copy Markdown

The latest updates on your projects. Learn more about Vercel for GitHub.

Project Deployment Actions Updated (UTC)
macromod Ready Ready Preview, Comment Jul 15, 2026 10:43pm

Request Review

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

1 participant