Skip to content

docs: refresh README to the actual repository and repair stale base pointers - #182

Merged
Aparnap2 merged 1 commit into
mainfrom
docs/readme-and-doc-pointers
Oct 24, 2026
Merged

Aparnap2 merged 1 commit into
mainfrom
docs/readme-and-doc-pointers

Conversation

@Aparnap2

Copy link
Copy Markdown
Owner

What changed

Four documentation files, no code:

File Change
README.md +27/−6 — the layout described a repository that no longer existed
docs/specs/06_PROGRESS.md +7/−1 — base pointer + the convention that stops the treadmill
docs/specs/09_DECISION_LOG.md +1/−1 — stale live pointer
docs/specs/10_PHASE3_E2E_QUALIFICATION.md +1/−1 — dated reconciliation narrative

Base main@c971727bf71a6a5dcedea9f426db0a8927a00bab · commit 163e205 · branch docs/readme-and-doc-pointers.

The README was stale — every claim below was verified against the tree

README said Reality (verified)
migrations 001 (schema+RLS) 002 (grants) 003 (evidence) 009 (investigations) Ten migrations, 001–010 (documents, outbox, worker role, investigation reports, app-core grants, workflow launches)
no tools/ entry tools/parsers/liteparse/shim.py is a production parse shim — exec'd by the Go adapter (client.go, 13 references) and copied into the worker image (Dockerfile.worker, 4 references) — so a Python runtime is required in production. tools/corpus/ is the sole reproducer of the committed corpus
no scripts/ci/ entry the four anti-fake-green + credential-isolation gates CI runs
no fixtures/parser_eval/ entry 45-case parser corpus, committed and sha256-verified
ADRs 001 002 004 008 009 also 005, 006, 007, and 011
tests/unit/ Python golden suite (no network, untouched by Go work) still true but understated — the Go coverage registry now asserts all 19 golden cases and fails a fixture case with no Go assertion
uv run pytest tests/ -q CI runs uv run pytest tests/unit -q; tests/ contains only unit/. Aligned, and the exact command was executed to confirm it works
Phase progression ended at "Next: Groq qualification" now records durable claim submission (#172), the 19-case contract, the expiry ADR (011) and the cleared advisory, and states plainly that release is NOT QUALIFIED

packages/domain/ is now labelled demoted per ADR-001, and tools/ is introduced with the heading "Python tooling that is NOT spec artifacts" so the distinction cannot be misread.

W17 — stale base pointers, history preserved

  • 09_DECISION_LOG.md — "current base is main@dffa160" → main@c971727.
  • 10_PHASE3_E2E_QUALIFICATION.md — the dated narrative keeps main@dffa160 (APA-64) as history and appends "Reconciled 2026-10-21: the base is now main@c971727."
  • 06_PROGRESS.md — pointer → main@c971727, plus a note recording the convention: the pointer records the commit whose content the file reflects; a documentation-only merge advances main afterwards, so the pointer trails by its own merge — that is expected, not drift. Without that note every documentation merge looks like a fresh regression, which is what kept W17 recurring.

The only surviving main@dffa160 anywhere in docs/ is that dated historical clause. No historical record was rewritten.

Verification

uv run pytest tests/unit -q   -> 67 passed        (the exact command the README now gives)
uv run ruff check .           -> All checks passed!
go build ./...                 -> clean
go test ./... -count=1        -> 41 packages ok, 0 failures

Every factual claim in the new README was re-derived from the tree, not from the old README.

Status

Documentation only. Not authorized to merge — opened as a draft for review. Release status NOT QUALIFIED; this closes no release gate and changes no behaviour.

@coderabbitai

coderabbitai Bot commented Oct 21, 2026 •

Copy link
Copy Markdown

Important

  • 🔍 Trigger review

This repository does not receive automatic reviews because it has fewer than 10 stars.

Run configuration
  • Source mode: Auto
  • Repository: Unavailable
  • Branch: Unavailable
  • File: No configuration file from this repository applied
  • Commit: Unavailable
  • Layers: defaults
  • Review profile: CHILL
  • Plan: Advanced
  • Run ID: ef91a20a-8754-4c4c-8b23-4cbe19984959
  • Autofix · Keep fixing CodeRabbit findings and required CI, and resolving merge conflicts

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

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

…ointers

The README described a repository that no longer exists, and was
omitted three load-bearing directories.

README corrections, each verified against the tree:
- Layout listed four migrations; there are ten (001-010). Now 001-010.
- Layout omitted tools/ entirely. tools/parsers/liteparse/shim.py is a
  PRODUCTION parse shim - executed by the Go adapter (client.go) and
  copied into the worker image - so a Python runtime is required in
  production, and tools/corpus/ is the sole reproducer of the committed
  corpus. Both are now listed, under a heading that distinguishes them
  from the spec artifacts.
- Layout omitted scripts/ci/, where the four anti-fake-green and
  credential-isolation gates CI runs live.
- Layout omitted fixtures/parser_eval/ (45 cases, sha256-verified).
- The ADR list omitted 005, 006, 007 and 011.
- 'Python golden suite (no network, untouched by Go work)' understated
  where the contract is enforced: the Go coverage registry now asserts
  all 19 golden cases and fails a fixture case that has no Go assertion.
- Setup said 'uv run pytest tests/ -q'; CI runs 'uv run pytest
  tests/unit -q'. Aligned, and the exact command was run to confirm it.
- Phase progression stopped at 'Next: Groq qualification'. Now records
  the durable claim submission, the 19-case contract, the expiry ADR and
  the cleared advisory, and states plainly that release is NOT QUALIFIED.

Stale base pointers repaired (W17):
- docs/specs/09_DECISION_LOG.md - 'current base is main@dffa160' ->
  main@c971727.
- docs/specs/10_PHASE3_E2E_QUALIFICATION.md - the dated reconciliation
  narrative keeps 'main@dffa160 (APA-64)' as history and appends the
  2026-10-21 reconciliation to main@c971727.
- docs/specs/06_PROGRESS.md - pointer -> main@c971727, with a note
  stating the convention: the pointer records the commit the file's
  content reflects, so a documentation-only merge advances main
  afterwards and the pointer trails by its own merge. That is expected,
  not drift.

Historical records were not rewritten: the only surviving main@dffa160
in docs/ is the dated historical clause inside the Phase-3 file.

Co-Authored-By: opencode <no-reply@opencode.ai>
@Aparnap2
Aparnap2 force-pushed the docs/readme-and-doc-pointers branch from 163e205 to 76a168d Compare October 24, 2026 05:46
@Aparnap2
Aparnap2 marked this pull request as ready for review October 24, 2026 05:49
@Aparnap2
Aparnap2 merged commit 2d3c57e into main Oct 24, 2026
7 checks passed
@Aparnap2
Aparnap2 deleted the docs/readme-and-doc-pointers branch October 24, 2026 05:49
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