v1.5.1 — polish: docs accuracy, test reliability, edge-case observability - #6
Conversation
Replace the example domain-folder name in README.md's negative-list narrative without changing scanner behavior.
- Replace concrete subcommand examples with stability guarantees - Preserve link to docs/JSON-SCHEMA.md and apiVersion mention - Removes release-blocker leaks ahead of v1.5.1
- Extend top-of-file block-comment with "Audit-trail invariant" subsection - Formalize the philosophical contract: "if it isn't journaled, it didn't happen" - Document why manifest_write_failed: failures count identically to disk-mutation failures in the all-failed gate - No production code changes; documentation-only addition
…ests - Add in-source tests asserting writeOp throw produces a failure record carrying the manifest_write_failed: prefix and counted in the failure tally (not success tally) - Add companion single-op test confirming all-failed gate fires identically for manifest_write_failed: as for disk-mutation failures - Locks Option (a): carving manifest_write_failed: out of the all-failed gate is now a regression
…uard; sync graceMs doc
…ith explicit narrowing
…emId(...) calls (A3)
Inserted '---' separator between the manifest-casing exception blockquote and the pre-dispatch validation envelope blockquote. Full-file scan found no additional MD028 instances.
…ite-number guard already present (C2)
Note: the file contains 4 it() cases, not 5 as the design source assumed. The header now mirrors reality — tests are the source of truth.
- Insert ## [1.5.1] - 2026-04-27 section above v1.5.0 header - Three subsections: Changed (README + _glyphs doc), Fixed (purge audit- trail, JSON-SCHEMA MD028, change-plan canonical-ID, scan-memory Windows paths), Tests (B1-B6 + Tier C fold-ins) - v1.5.0 entry preserved byte-for-byte
|
Warning Rate limit exceeded
To keep reviews running without waiting, you can enable usage-based add-on for your organization. This allows additional reviews beyond the hourly cap. Account admins can enable it under billing. ⌛ How to resolve this issue?After the wait time has elapsed, a review can be triggered using the We recommend that you space out your commits to avoid hitting the rate limit. 🚦 How do rate limits work?CodeRabbit enforces hourly rate limits for each developer per organization. Our paid plans have higher rate limits than the trial, open-source and free plans. In all cases, we re-allow further reviews after a brief timeout. Please see our FAQ for further information. 📝 WalkthroughWalkthroughPatch release v1.5.1: documentation and README clarifications, test reliability and alignment fixes, small UI/formatting tweaks, replacement of hard-coded canonical-id literals in tests with helper calls, and added regression tests (including Windows path-normalization and purge audit-trail scenarios). Changes
Estimated code review effort🎯 3 (Moderate) | ⏱️ ~25 minutes Poem
🚥 Pre-merge checks | ✅ 5✅ Passed checks (5 passed)
✏️ Tip: You can configure your own custom pre-merge checks in the settings. ✨ Finishing Touches🧪 Generate unit tests (beta)
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. Comment |
![coderabbitai[bot]](https://avatars.githubusercontent.com/in/347564?s=60&v=4){kind=small}
There was a problem hiding this comment.
Actionable comments posted: 2
Caution
Some comments are outside the diff and can’t be posted inline due to platform limitations.
⚠️ Outside diff range comments (1)
README.md (1)
34-34:⚠️ Potential issue | 🟡 MinorUpdate README "Current release" and "Current package version" to 1.5.1.
The CHANGELOG adds a
## [1.5.1]entry andapps/ccaudit/package.jsonis at1.5.1, but README.md still advertises v1.5.0 at line 34 and line 686. Update both so users see the released version.📝 Suggested fixes
Line 34:
-Current release: **v1.5.0** — interactive archive picker, interactive restore picker, fuzzy-match restore by name, archive purge. See [CHANGELOG.md](./CHANGELOG.md). +Current release: **v1.5.1** — interactive archive picker, interactive restore picker, fuzzy-match restore by name, archive purge. See [CHANGELOG.md](./CHANGELOG.md).Line 686:
-- Current package version: **1.5.0** +- Current package version: **1.5.1**🤖 Prompt for AI Agents
Verify each finding against the current code and only fix it if needed. In `@README.md` at line 34, Update the two occurrences of the displayed version in README: change the "Current release: **v1.5.0**" text (the "Current release" line) and the "Current package version" value (the "Current package version" line) from v1.5.0 to v1.5.1 so the README matches the changelog and apps/ccaudit/package.json; search for the literal "v1.5.0" or the headings "Current release" and "Current package version" and replace with "v1.5.1".
🧹 Nitpick comments (5)
apps/ccaudit/src/__tests__/fixtures/tmux-e2e.ts (1)
145-168: LGTM — redundant strip removed, invariant preserved.The mapping
ansi: opts.stripAnsi === falsecorrectly delegates ANSI handling tocapture():
stripAnsiunset/true→ansi: false→capture()returns stripped output.stripAnsi: false→ansi: true→capture()returns raw bytes.So
lastCapture = rawis in the desired shape in both branches and matching/error reporting are unaffected.Optional nit: the local name
rawis a little misleading in the default path (wherecapture()has already stripped ANSI). Renaming to something likecapturedwould more accurately reflect the dual-state semantics, but it's purely cosmetic.🤖 Prompt for AI Agents
Verify each finding against the current code and only fix it if needed. In `@apps/ccaudit/src/__tests__/fixtures/tmux-e2e.ts` around lines 145 - 168, The code in waitForText correctly delegates ANSI handling to capture, but the local variable name raw is misleading because capture may already strip ANSI; rename raw to captured (and update its usage where assigned to lastCapture and where it's declared) and adjust the adjacent comment to reflect that capture returns either stripped or raw output depending on opts.stripAnsi (via ansi: opts.stripAnsi === false), leaving all logic and the matching/error message unchanged; update references to raw, lastCapture assignment, and the comment inside waitForText accordingly.apps/ccaudit/scripts/bundle-size-check.mjs (2)
17-26: Minor formatting inconsistency informatBytesnon-finite fallback.The non-finite branch returns
${bytes}B(no space) while the byte branch on line 25 returns${bytes} B(with space). Not currently reachable forBUDGET_BYTES, but worth normalizing to keep output uniform.♻️ Optional refactor
- if (!Number.isFinite(bytes)) return `${bytes}B`; + if (!Number.isFinite(bytes)) return `${bytes} B`;🤖 Prompt for AI Agents
Verify each finding against the current code and only fix it if needed. In `@apps/ccaudit/scripts/bundle-size-check.mjs` around lines 17 - 26, The non-finite branch in function formatBytes returns `${bytes}B` (no space) while other branches use a space before "B"/"KB"/"MB"; update the non-finite return to use a space (e.g. `${bytes} B`) so output formatting is consistent in formatBytes, and ensure the change is applied only inside the formatBytes function to normalize spacing across all branches.
96-99: Optional: applyformatBytesto phase-local FAIL message too.The primary FAIL message on line 60 now reports a human-readable budget, but the phase-local FAIL on line 98 still emits raw
${phaseBudget}B. UsingformatBytes(phaseBudget)here keeps the messages consistent.♻️ Optional refactor
- `[bundle-size] FAIL: phase-local delta exceeds ${phaseBudget}B (${phaseDelta} > ${phaseBudget})`, + `[bundle-size] FAIL: phase-local delta exceeds ${formatBytes(phaseBudget)} (${phaseDelta} > ${phaseBudget})`,🤖 Prompt for AI Agents
Verify each finding against the current code and only fix it if needed. In `@apps/ccaudit/scripts/bundle-size-check.mjs` around lines 96 - 99, The phase-local FAIL log uses raw bytes ("${phaseBudget}B") instead of the human-readable formatter; update the console.error in the branch that checks if (phaseDelta > phaseBudget) to use formatBytes(phaseBudget) (and keep phaseDelta as-is or format if desired) so messages match the primary FAIL format—look for the variables phaseDelta, phaseBudget and the formatBytes function in bundle-size-check.mjs and replace the raw `${phaseBudget}B` insertion with formatBytes(phaseBudget).CHANGELOG.md (1)
36-53: Minor: categorization mismatch in### Fixedvs### Tests.The
scan-memory.tsbullet (lines 36–38) explicitly notes the normalization "was already correct; the test locks it" — i.e., it's regression coverage, not a behavior fix. Since you've already introduced a dedicated### Testssection directly below, moving this entry there would be more consistent and avoid implying a user-visible fix landed.Also,
### Testsisn't part of the Keep a Changelog 1.1.0 category set referenced at the top of the file (Added/Changed/Deprecated/Removed/Fixed/Security). Not a blocker — many projects extend the schema — but worth a thought if you want to stay strictly conformant.📝 Suggested move
### Fixed @@ - `change-plan.ts`: replaced hard-coded canonical-ID literal strings with calls to the exported `canonicalItemId(...)` helper, eliminating drift risk if the format string changes. -- `scan-memory.ts`: added Windows path-normalization regression coverage - (in-source test). The normalization itself was already correct; the - test locks it. ### Tests +- `scan-memory.ts`: added Windows path-normalization regression coverage + (in-source test). The normalization itself was already correct; the + test locks it. - Test-helper polish: replaced a dead `void killed` no-op with a real🤖 Prompt for AI Agents
Verify each finding against the current code and only fix it if needed. In `@CHANGELOG.md` around lines 36 - 53, The changelog places the `scan-memory.ts` bullet under `### Fixed` but it describes regression test coverage, so move that bullet into the existing `### Tests` section (or change its phrasing to explicitly say "test/regression coverage" rather than a fix); update the `scan-memory.ts` line to read that it adds Windows path-normalization regression coverage (in-source test) and remove any wording that implies a user-visible fix; optionally, if you want strict Keep a Changelog conformance, replace or omit the `### Tests` heading and instead add the bullet under an appropriate official category (e.g., `Changed` or a dedicated testing note) so the entry's intent matches its section.README.md (1)
704-718: LGTM — abstract contract framing is the right call for an unreleased v1.6.Removing the concrete
ccaudit game ...subcommand list and the externalcommand: "game.*"envelope guarantee in favor ofapiVersionframing avoids prematurely committing to a surface that isn't shipped yet. The "not part of the contract" negative bullets (TUI, tables, env vars, internal TS, path-shaped canonical IDs) read as a clear stability boundary.One small thought: the section header dropped "game" but the surrounding paragraph doesn't say what is covered (commands? envelopes? both?). If the v1.6 release notes aren't ready yet, consider a one-line "scope: report commands'
--jsonoutput" hint so embedders reading this today have a rough mental model. Optional.🤖 Prompt for AI Agents
Verify each finding against the current code and only fix it if needed. In `@README.md` around lines 704 - 718, The header "v1.6 stable CLI/JSON API contract" and paragraph mention `apiVersion` and `--json` but don't state what is in-scope; add a single clarifying line under that paragraph such as "Scope: report commands' `--json` output (CLI subcommands' JSON envelopes including `apiVersion` must adhere to docs/JSON-SCHEMA.md)" so readers know the contract covers command JSON envelopes; reference the header "v1.6 stable CLI/JSON API contract", the `--json` flag, `apiVersion`, and docs/JSON-SCHEMA.md when adding this line.
🤖 Prompt for all review comments with AI agents
Verify each finding against the current code and only fix it if needed.
Inline comments:
In `@apps/ccaudit/src/__tests__/_test-helpers.ts`:
- Around line 10-18: The docstring is inaccurate: update it to state that the
poll loop is bounded by the hard-coded earlyExitMs (earlyExitMs = 1000) rather
than graceMs, and that graceMs is only the final configurable sleep added after
the poll completes (default 300ms), not the duration that the poll runs;
reference the variables earlyExitMs and graceMs in the docstring and remove the
contradictory phrase that the poll runs until graceMs then a 300ms grace is
added.
In `@README.md`:
- Line 621: Add "data-engineering" to the DOMAIN_STOP_FOLDERS set so the
explicit Tier 2 stop-list matches the README's claim; update the corresponding
unit test that asserts the stop-list length (and expected entries) from 18 to 19
to include the new "data-engineering" entry. Modify the DOMAIN_STOP_FOLDERS
constant and the test that validates its count/contents (the test referencing
DOMAIN_STOP_FOLDERS) so they both include "data-engineering" to restore dual
enforcement described in the README.
---
Outside diff comments:
In `@README.md`:
- Line 34: Update the two occurrences of the displayed version in README: change
the "Current release: **v1.5.0**" text (the "Current release" line) and the
"Current package version" value (the "Current package version" line) from v1.5.0
to v1.5.1 so the README matches the changelog and apps/ccaudit/package.json;
search for the literal "v1.5.0" or the headings "Current release" and "Current
package version" and replace with "v1.5.1".
---
Nitpick comments:
In `@apps/ccaudit/scripts/bundle-size-check.mjs`:
- Around line 17-26: The non-finite branch in function formatBytes returns
`${bytes}B` (no space) while other branches use a space before "B"/"KB"/"MB";
update the non-finite return to use a space (e.g. `${bytes} B`) so output
formatting is consistent in formatBytes, and ensure the change is applied only
inside the formatBytes function to normalize spacing across all branches.
- Around line 96-99: The phase-local FAIL log uses raw bytes ("${phaseBudget}B")
instead of the human-readable formatter; update the console.error in the branch
that checks if (phaseDelta > phaseBudget) to use formatBytes(phaseBudget) (and
keep phaseDelta as-is or format if desired) so messages match the primary FAIL
format—look for the variables phaseDelta, phaseBudget and the formatBytes
function in bundle-size-check.mjs and replace the raw `${phaseBudget}B`
insertion with formatBytes(phaseBudget).
In `@apps/ccaudit/src/__tests__/fixtures/tmux-e2e.ts`:
- Around line 145-168: The code in waitForText correctly delegates ANSI handling
to capture, but the local variable name raw is misleading because capture may
already strip ANSI; rename raw to captured (and update its usage where assigned
to lastCapture and where it's declared) and adjust the adjacent comment to
reflect that capture returns either stripped or raw output depending on
opts.stripAnsi (via ansi: opts.stripAnsi === false), leaving all logic and the
matching/error message unchanged; update references to raw, lastCapture
assignment, and the comment inside waitForText accordingly.
In `@CHANGELOG.md`:
- Around line 36-53: The changelog places the `scan-memory.ts` bullet under `###
Fixed` but it describes regression test coverage, so move that bullet into the
existing `### Tests` section (or change its phrasing to explicitly say
"test/regression coverage" rather than a fix); update the `scan-memory.ts` line
to read that it adds Windows path-normalization regression coverage (in-source
test) and remove any wording that implies a user-visible fix; optionally, if you
want strict Keep a Changelog conformance, replace or omit the `### Tests`
heading and instead add the bullet under an appropriate official category (e.g.,
`Changed` or a dedicated testing note) so the entry's intent matches its
section.
In `@README.md`:
- Around line 704-718: The header "v1.6 stable CLI/JSON API contract" and
paragraph mention `apiVersion` and `--json` but don't state what is in-scope;
add a single clarifying line under that paragraph such as "Scope: report
commands' `--json` output (CLI subcommands' JSON envelopes including
`apiVersion` must adhere to docs/JSON-SCHEMA.md)" so readers know the contract
covers command JSON envelopes; reference the header "v1.6 stable CLI/JSON API
contract", the `--json` flag, `apiVersion`, and docs/JSON-SCHEMA.md when adding
this line.
🪄 Autofix (Beta)
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: defaults
Review profile: CHILL
Plan: Pro
Run ID: 4274f389-93d5-42e7-ad51-6b369a6aa845
📒 Files selected for processing (16)
CHANGELOG.mdREADME.mdapps/ccaudit/package.jsonapps/ccaudit/scripts/bundle-size-check.mjsapps/ccaudit/src/__tests__/_test-helpers.tsapps/ccaudit/src/__tests__/fixtures/tmux-e2e.tsapps/ccaudit/src/__tests__/pagination-500.test.tsapps/ccaudit/src/__tests__/restore-corrupt-manifest.test.tsapps/ccaudit/src/__tests__/restore-json-envelope.test.tsdocs/JSON-SCHEMA.mdpackages/internal/src/remediation/change-plan.tspackages/internal/src/remediation/purge.tspackages/internal/src/scanner/scan-memory.tspackages/terminal/src/tables/change-plan.tspackages/terminal/src/tui/_force-partial-banner.tspackages/terminal/src/tui/_glyphs.ts
| /** | ||
| * Wait for the spawned picker subprocess to reach its blocking read loop. | ||
| * Polls until the child exits (error/crash path), or until maxWaitMs elapses, | ||
| * Polls until the child exits (error/crash path), or until graceMs elapses, | ||
| * with exponential backoff — whichever comes first. Then adds a final grace | ||
| * delay so the TUI is ready for key input. | ||
| * | ||
| * Typical path: child never exits during the poll window, loop runs until | ||
| * maxWaitMs, then a 300ms grace is added for the render cycle to settle. | ||
| * graceMs, then a 300ms grace is added for the render cycle to settle. | ||
| */ |
There was a problem hiding this comment.
Docstring still mismatches the implementation — graceMs does not bound the poll loop.
The rewritten docstring conflates two distinct timeouts. In the implementation:
- The poll loop is bounded by the hard-coded
earlyExitMs = 1_000(lines 21, 24), notgraceMs. graceMsis only the final sleep added once the poll loop completes without an early exit (lines 29–31), and it is configurable via the parameter (default 300ms), not a fixed 300ms.
So both phrases — "Polls until … graceMs elapses" and "loop runs until graceMs, then a 300ms grace is added" — are inaccurate. Given that doc accuracy is one of this PR's stated goals, worth tightening here.
📝 Suggested docstring
/**
* Wait for the spawned picker subprocess to reach its blocking read loop.
- * Polls until the child exits (error/crash path), or until graceMs elapses,
- * with exponential backoff — whichever comes first. Then adds a final grace
- * delay so the TUI is ready for key input.
+ * Polls (with exponential backoff) for up to ~1s to detect an early exit
+ * (error/crash path). If the child is still alive after that window, sleeps
+ * an additional `graceMs` so the TUI has time to reach its blocking read.
*
- * Typical path: child never exits during the poll window, loop runs until
- * graceMs, then a 300ms grace is added for the render cycle to settle.
+ * Typical path: the child does not exit during the ~1s poll window, then
+ * `graceMs` (default 300ms) is added for the render cycle to settle.
*/📝 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.
| /** | |
| * Wait for the spawned picker subprocess to reach its blocking read loop. | |
| * Polls until the child exits (error/crash path), or until maxWaitMs elapses, | |
| * Polls until the child exits (error/crash path), or until graceMs elapses, | |
| * with exponential backoff — whichever comes first. Then adds a final grace | |
| * delay so the TUI is ready for key input. | |
| * | |
| * Typical path: child never exits during the poll window, loop runs until | |
| * maxWaitMs, then a 300ms grace is added for the render cycle to settle. | |
| * graceMs, then a 300ms grace is added for the render cycle to settle. | |
| */ | |
| /** | |
| * Wait for the spawned picker subprocess to reach its blocking read loop. | |
| * Polls (with exponential backoff) for up to ~1s to detect an early exit | |
| * (error/crash path). If the child is still alive after that window, sleeps | |
| * an additional `graceMs` so the TUI has time to reach its blocking read. | |
| * | |
| * Typical path: the child does not exit during the ~1s poll window, then | |
| * `graceMs` (default 300ms) is added for the render cycle to settle. | |
| */ |
🤖 Prompt for AI Agents
Verify each finding against the current code and only fix it if needed.
In `@apps/ccaudit/src/__tests__/_test-helpers.ts` around lines 10 - 18, The
docstring is inaccurate: update it to state that the poll loop is bounded by the
hard-coded earlyExitMs (earlyExitMs = 1000) rather than graceMs, and that
graceMs is only the final configurable sleep added after the poll completes
(default 300ms), not the duration that the poll runs; reference the variables
earlyExitMs and graceMs in the docstring and remove the contradictory phrase
that the poll runs until graceMs then a 300ms grace is added.
| > `marketing/`, `testing/`, `sales/`, `integrations/`, `strategy/`, | ||
| > `project-management/`, `support/`, `paid-media/`, `spatial-computing/`, | ||
| > `examples/`, `scripts/`, `product/`, `specialized/`, `game-development/`, | ||
| > `examples/`, `scripts/`, `product/`, `specialized/`, `data-engineering/`, |
There was a problem hiding this comment.
🧩 Analysis chain
🏁 Script executed:
#!/bin/bash
# Confirm DOMAIN_STOP_FOLDERS contains data-engineering to match README claim.
rg -nP -C2 'DOMAIN_STOP_FOLDERS' --type=ts
echo "--- searching for data-engineering literal ---"
rg -nP "['\"]data-engineering['\"]" --type=tsRepository: fabio-dee/ccaudit
Length of output: 11153
🏁 Script executed:
# Read the complete DOMAIN_STOP_FOLDERS definition from stop-lists.ts
sed -n '54,85p' packages/internal/src/framework/stop-lists.tsRepository: fabio-dee/ccaudit
Length of output: 852
🏁 Script executed:
# Read the README context around line 621 to understand the full claim
sed -n '610,630p' README.mdRepository: fabio-dee/ccaudit
Length of output: 1168
data-engineering/ is listed in README but missing from DOMAIN_STOP_FOLDERS.
The README (line 615–627) claims data-engineering/ is a domain folder that "ccaudit refuses to group," and explicitly states this is "enforced twice: once by gating folder-segment matches through the curated list only (Tier 1), and once by an explicit DOMAIN_STOP_FOLDERS list."
However, DOMAIN_STOP_FOLDERS in packages/internal/src/framework/stop-lists.ts (lines 54–71) contains 18 entries but does not include data-engineering. The test confirming the count and validating the expected list does not account for this entry either. This breaks the promised dual enforcement: Tier 2 will not prevent data-engineering/ from being auto-promoted to a framework if items cluster by prefix.
Add data-engineering to the DOMAIN_STOP_FOLDERS set and update the test count from 18 to 19.
🤖 Prompt for AI Agents
Verify each finding against the current code and only fix it if needed.
In `@README.md` at line 621, Add "data-engineering" to the DOMAIN_STOP_FOLDERS set
so the explicit Tier 2 stop-list matches the README's claim; update the
corresponding unit test that asserts the stop-list length (and expected entries)
from 18 to 19 to include the new "data-engineering" entry. Modify the
DOMAIN_STOP_FOLDERS constant and the test that validates its count/contents (the
test referencing DOMAIN_STOP_FOLDERS) so they both include "data-engineering" to
restore dual enforcement described in the README.
…shes from v1.5.1 CHANGELOG entry
There was a problem hiding this comment.
Caution
Some comments are outside the diff and can’t be posted inline due to platform limitations.
⚠️ Outside diff range comments (2)
README.md (2)
711-715:⚠️ Potential issue | 🟡 MinorVersion mismatch: “Current package version: 1.5.0” vs release v1.5.1.
Near the end, the README lists:
- Current release: v1.5.1 (Line 35)
- Current package version: 1.5.0 (Line 713)
Unless that’s intentionally documenting something else, this is likely stale and confusing for users.
Consider updating the README version stanza to 1.5.1 (or clarifying why it remains 1.5.0).
🤖 Prompt for AI Agents
Verify each finding against the current code and only fix it if needed. In `@README.md` around lines 711 - 715, Update the README stanza that shows "Current package version: **1.5.0**" to match the release v1.5.1 (or add a clarifying note if the package intentionally remains at 1.5.0); locate the "Current package version" text in the README (the block containing "Current package version: **1.5.0**" and the referenced `apps/ccaudit/package.json` and `apps/ccaudit/src/_version.ts`) and change the displayed version to **1.5.1** (or append a brief explanation why it differs).
643-655:⚠️ Potential issue | 🟠 MajorFix doc/code contract mismatch:
data-engineering/is not inDOMAIN_STOP_FOLDERS;game-development/is missing from the list.The README lists
data-engineering/as enforced by the Tier 2DOMAIN_STOP_FOLDERSconstant, but the actual set atpackages/internal/src/framework/stop-lists.ts:54contains 18 entries excludingdata-engineering/. Instead, the set includesgame-development/, which is not mentioned in the README.Update the README list to match the actual 18 entries in the code:
- Remove:
data-engineering/- Add:
game-development/🤖 Prompt for AI Agents
Verify each finding against the current code and only fix it if needed. In `@README.md` around lines 643 - 655, The README's list of domain stop folders is out of sync with the code: update the README entry under "Domain folders are NOT frameworks" so it matches the actual constant DOMAIN_STOP_FOLDERS used by the framework logic — remove "data-engineering/" and add "game-development/" to mirror the 18 entries defined in the code (refer to the DOMAIN_STOP_FOLDERS constant) so documentation and implementation are consistent.
🤖 Prompt for all review comments with AI agents
Verify each finding against the current code and only fix it if needed.
Outside diff comments:
In `@README.md`:
- Around line 711-715: Update the README stanza that shows "Current package
version: **1.5.0**" to match the release v1.5.1 (or add a clarifying note if the
package intentionally remains at 1.5.0); locate the "Current package version"
text in the README (the block containing "Current package version: **1.5.0**"
and the referenced `apps/ccaudit/package.json` and
`apps/ccaudit/src/_version.ts`) and change the displayed version to **1.5.1**
(or append a brief explanation why it differs).
- Around line 643-655: The README's list of domain stop folders is out of sync
with the code: update the README entry under "Domain folders are NOT frameworks"
so it matches the actual constant DOMAIN_STOP_FOLDERS used by the framework
logic — remove "data-engineering/" and add "game-development/" to mirror the 18
entries defined in the code (refer to the DOMAIN_STOP_FOLDERS constant) so
documentation and implementation are consistent.
ℹ️ Review info
⚙️ Run configuration
Configuration used: defaults
Review profile: CHILL
Plan: Pro
Run ID: ab26c31a-7a9d-4829-969a-1fdb7c7226c9
📒 Files selected for processing (2)
CHANGELOG.mdREADME.md
✅ Files skipped from review due to trivial changes (1)
- CHANGELOG.md
…and bump package-version label to 1.5.1 Addresses CodeRabbit findings on PR #6: - Stop-folder list cited data-engineering/ but DOMAIN_STOP_FOLDERS in packages/internal/src/framework/stop-lists.ts contains game-development/. - Version stanza near footer still showed 1.5.0; release is 1.5.1.
1 participant
Summary
Polish release. No user-visible behavior changes. No safety invariants (INV-S1..INV-S6) touched. No interactive picker TUI files modified.
## [1.5.1]entry added.packages/internal/src/remediation/purge.tsdocuments the "if it isn't journaled, it didn't happen" contract and locks it with in-source tests againstmanifest_write_failed:regressions.Scope
b26ae7e) and HEAD onrelease/v1.5.1.pnpm verifygreen (typecheck + lint + build + 1788 tests + format:check).packages/internal/src/remediation/{bust,restore,reclaim,manifest,checkpoint}.ts— UNTOUCHED.Test plan
pnpm verifygreen on CI## [1.5.1]entry rendered correctlySummary by CodeRabbit
New Features
Documentation
Bug Fixes
Chores