[Phase 16] S15 — Documentation: claims bound to code
Nobody has claimed this yet.
Assessment
- Difficulty
- 4/5
- Estimated time
- 3-5 days
- Newbie friendliness
- 45/100
- Issue type
- Documentation
- Clarity
- Needs clarification
- Activity status
- Active
- Tech stack
- html, python
- Domain
- documentation
Research direction
Read the roadmap section at docs/roadmap/phase-16-trust-surface-hardening.md lines 746-795 and the existing auto-revert.md page. Locate the eight auto-revert claim instances across the codebase and the six site locales, then review check_cli_exit_code_prose.py to understand the guard pattern. Run ./.venv/bin/python to execute the test gauntlet. Done when all exit criteria pass: no outdated legal dates, no restore claims in surfaces, exit-code prose matches _PUBLIC_EXIT_CODES, and all guards are green.
Written by the indexing model from the issue text.
Description
Roadmap · Phase 16, step S15 · size XL · 10 unit(s) · planned in
docs/roadmap/phase-16-trust-surface-hardening.md:746-795
Summary
Published claims the code does not back, plus the guards that would have caught them. Regulation (EU) 2026/1744 moves Annex III high-risk obligations to 2 December 2027 and Annex I to 2 August 2028, while current surfaces still tie "full enforcement" to 2 August 2026. The auto-revert "rollback" claim survives in eight places across nine files under four wordings — _revert_model deletes and restores nothing. EN, TR and all six site locales change in the same commit. (DOCS-06 is a residual wording fix only; its main claim was refuted.)
Units
| Unit | Tracked by (October 2026 review) |
|---|---|
F-W20260729-DOCS-01 |
#75 · #748 · #806 |
F-W20260729-DOCS-02 |
#87 · #89 · #128 · #280 · #282 · #284 · #291 · #360 · #367 · #453 · #458 · #459 · #519 · #540 · #556 · #557 · #558 · #599 · #683 · #684 · #685 |
F-W20260729-DOCS-04 |
#114 · #208 · #345 · #570 · #576 · #580 · #583 · #594 · #615 · #616 · #622 · #624 · #633 · #684 · #710 · #725 · #734 · #766 · #772 · #777 · #784 · #790 · #791 · #820 |
F-W20260729-DOCS-03 |
#114 · #208 · #367 · #576 · #580 · #583 · #594 · #684 · #725 · #734 · #766 · #777 |
C31-EXIT-CODE-PROSE (= DOCS-07 + OPS-24) |
#87 · #150 · #162 · #183 · #369 · #445 · #447 · #454 · #543 · #547 · #558 · #580 · #583 · #591 · #604 · #607 · #633 · #654 · #655 · #657 · #668 · #670 · #671 · #676 · #680 · #681 · #686 · #715 · #719 · #721 · #742 · #748 · #766 · #774 · #776 |
C31-DOC-SCHEMA-PROSE (= DOCS-08 + DOCS-09) |
#75 · #101 · #147 · #199 · #228 · #229 · #230 · #231 · #237 · #248 · #257 · #358 · #359 · #424 · #425 · #426 · #427 · #452 · #503 · #545 · #546 · #550 · #558 · #569 · #589 · #593 · #615 · #624 · #625 · #627 · #641 · #642 · #662 · #663 · #670 · #671 · #677 · #689 · #791 · #801 · #813 |
F-W20260729-DOCS-14, F-W20260729-DOCS-15, F-W20260729-DOCS-16 |
#128 · #360 · #367 · #453 · #458 · #459 · #519 · #540 · #556 · #557 · #558 · #598 · #683 · #684 · #685 · #732 · #770 · #771 · #797 · #799 |
F-W20260729-DOCS-06 |
#366 |
Decisions this step executes
- GTM — AI Act moat framing after Regulation (EU) 2026/1744: retarget to 2 December 2027, framing preserved.
Plan
- Commit 1: legal dates + a freshness guard. Commit 2: the auto-revert semantics sweep + a semantic guard. Commit 3: site runtime claims across six locales, static fallbacks, privacy / localStorage, a11y and assets. Commit 4: exit-code prose, ghost config keys, the staging path and guard extensions.
docs/roadmap/risks-and-decisions.md's append-only Decision Log is not a target; its live risk rows are.- Extend
check_cli_exit_code_prose.pynon-strict first, triage the baseline, then flip.
Exit criteria
- No current surface ties high-risk obligations to August 2026, and a guard fails on reintroduction.
- No locale claims an unconditional exit 3 or a generated conformity declaration, verified in a rendered-DOM-equivalent test for EN plus at least one other locale.
- A grep for restore claims over current surfaces returns zero (the canonical
auto-revert.mdpage is exempt because it quotes the wrong claim to refute it). - A localStorage snapshot after every wizard step contains no credential-class value.
- Every current surface states
verify-auditas 0/1/2/6 with the "1 = unverified, 6 = compared and failed" distinction, and a meta-test scans the canonical exit-code statements in all four rulebooks against_PUBLIC_EXIT_CODES. - An inline-code config key in prose that does not resolve against
ForgeConfigfails the guard. -
aria-expandedtracks drawer state across all four flows, and every asset referenced fromsite/*.htmlresolves in the deploy source tree.
How to deliver
- One root-cause family, committed on its own; then an Opus review round and a Sonnet review round, each with verified findings fixed and committed before the next step begins.
- A reproducer that is red before the fix and green after; the full gauntlet from
CLAUDE.mdpasses under./.venv/bin/python. - Every English documentation edit ships with its Turkish mirror in the same commit; a new or renamed audit event gets its EN and TR catalog rows in the same PR.
- A step that grows a deferred module carries its
budget_historyjustification in the same diff; a widened guard rolls out non-strict first, then flips. - Commit bodies cite
Absorbs F-W20260729-…; the CHANGELOG[Unreleased]section gets an entry only for a change a user or library consumer can observe (decision C-20).
Planned in docs/roadmap/phase-16-trust-surface-hardening.md (read at f94595f). Unit IDs (F-W20260729-…) come from the 2026-07-29/30 full-project review; the issue numbers next to them are the October 2026 review's records of the same defects.
- Dominant language
- Python
- Stars
- 9
- Forks
- 1
- Avg merge
- 8h 10m
- Merged PRs (30d)
- 5
Getting set up
- Ships a Dockerfile or Docker Compose file
- Has a pull request template
- Read the contributing guide
First steps
- Read the whole issue, then the project's contributing guide.
- Comment on the issue to say you are picking it up — it saves two people doing the same work.
- Fork the repository and make your change on a branch.
- Open a pull request that references the issue number.
More from HodeTech/ForgeLM
-
area: dev-tooling bug severity: low source: roadmap wave: 4
Difficulty 2/5 1-3 hours Newbie friendliness 70/100
-
Site: the theme-locked --code-comment grey (#6a6f80) on the dark code surface is 3.8:1, below WCAG AA, in both themes (F-W1-SITE-06)Possibly taken @vortsghost2025 claimed this today. Openarea: site bug good first issue severity: low source: review-2026-09 wave: 4
Difficulty 2/5 1-3 hours Newbie friendliness 66/100
-
documentation good first issue severity: medium source: review-2026-09 wave: 3
Difficulty 2/5 1-3 hours Newbie friendliness 75/100
-
documentation good first issue severity: low source: review-2026-09 wave: 4
Difficulty 1/5 1-3 hours Newbie friendliness 80/100
-
documentation severity: medium source: review-2026-09 wave: 3
Difficulty 2/5 1-3 hours Newbie friendliness 72/100
All issues in HodeTech/ForgeLM
Similar issues
-
Difficulty 2/5 1-3 hours Newbie friendliness 72/100
Maintainers usually reply within 1 day
-
task
Difficulty 2/5 1-3 hours Newbie friendliness 78/100
danparshall/claude_researcher#87 ·
Maintainers usually reply within 1 day
-
bug
Difficulty 2/5 1-3 hours Newbie friendliness 85/100
datajoint/datajoint-python#1586 ·
Maintainers usually reply within 1 day
-
1D intercept of length 1 is accepted as `fix_params` in population GLM with more than 1 unitPossibly taken @dhvani7014 claimed this today. Openbug good first issue
Difficulty 2/5 1-3 hours Newbie friendliness 72/100
flatironinstitute/nemos#642 · 1 comment ·
Maintainers usually reply within 1 day
-
Difficulty 1/5 Under an hour Newbie friendliness 78/100
uniswap-python/uniswap-python#487 ·