fix(signals): align focus-manifest.ts's glob cap with group-based counting

[JSONbored]

Summary

• focus-manifest.ts's normalizeOptionalGlob (parses contentLane.entryFileGlob/providerFileGlob/artifactGlob from .gittensory.yml) capped raw * character count at 3 — the same flawed model change-guardrail.ts's globToRegExp/matchesAny used before fix(signals): cap globToRegExp wildcard count to prevent ReDoS #2445's fix. A ** pair compiles to a single .* group, not two independent wildcards, so raw character counting is wrong in both directions: it wrongly rejects a legitimate 2-group glob like "public/**/*.json" (3 raw star characters) in some cases, and would wrongly admit a genuinely dangerous 3-single-star-group glob (e.g. "a-*-*-*-b") at the same raw count — exactly the class of bug fix(signals): cap globToRegExp wildcard count to prevent ReDoS #2445 fixed.
• Not a live security hole. globToRegExp (from fix(signals): cap globToRegExp wildcard count to prevent ReDoS #2445) already has its own internal group-based cap (MAX_GLOB_WILDCARD_GROUPS = 2) that safely short-circuits any over-complex glob to a never-matching sentinel regex regardless of what this parse-time check lets through — full defense-in-depth already exists. This closes a consistency/UX gap: today some over-complex contentLane.*Glob values get an explicit "too many wildcards" parse-time warning while others silently pass validation and then silently compile to a never-matching pattern at match-time with zero warning.
• Fix: export change-guardrail.ts's countWildcardGroups/MAX_GLOB_WILDCARD_GROUPS (both already benchmarked and battle-tested in fix(signals): cap globToRegExp wildcard count to prevent ReDoS #2445) and reuse them directly in focus-manifest.ts instead of re-deriving a second, separately-maintained threshold — the two glob-safety checks in this codebase now share one source of truth instead of risking future drift. change-guardrail.ts has zero imports of its own (confirmed dependency-free), so this is a safe, non-circular sibling import within src/signals/.

Scope

• The PR title follows type(scope): short summary Conventional Commit format.
• This PR is focused (glob-cap consistency only) and does not mix unrelated backend, UI, MCP, docs, dependency, and deploy changes.
• This follows CONTRIBUTING.md and does not reintroduce GitHub Pages, VitePress, site/, or CNAME.
• I linked an issue, or this is small enough that the summary explains why an issue is not needed. (No issue — a small, self-contained follow-up flagged during fix(signals): cap globToRegExp wildcard count to prevent ReDoS #2445's review, same class of fix.)

Validation

• git diff --check
• npm run actionlint
• npm run typecheck
• npm run test:coverage locally — the one pre-existing sub-100% line in focus-manifest.ts (MAX_FOCUS_MANIFEST_BYTES over-length content path) is untouched by this diff and unrelated; every line/branch this PR changes is covered
• npm run test:workers
• npm run build:mcp
• npm run test:mcp-pack
• npm run ui:openapi:check
• npm run ui:lint
• npm run ui:typecheck
• npm run ui:build
• npm audit --audit-level=moderate
• New or changed behavior has unit/integration tests for new branches, fallback paths, and sanitizer boundaries — added a REGRESSION test proving "public/**/*.json" (2 groups, 3 raw characters) is now accepted, and updated the existing "too many wildcards" test's "at cap" example to the new 2-group boundary

Safety

• No secrets, wallet details, hotkeys, coldkeys, user PATs, private keys, raw trust scores, private rankings, or private maintainer evidence are exposed.
• Public GitHub text stays sanitized, low-noise, and does not imply compensation guarantees or optimization tactics.
• Auth, cookie, CORS, GitHub App, Cloudflare, or session changes include negative-path tests. (N/A — no such changes.)
• API/OpenAPI/MCP behavior is updated and tested where needed. (N/A — internal config-parsing helper only.)
• UI changes use live API data or real empty/error/loading states. (N/A — no UI changes.)
• Visible UI changes include a UI Evidence section. (N/A — no UI changes.)
• Public docs/changelogs are updated where needed. (N/A — no public-facing behavior change; this is an internal consistency fix.)

Notes

• This targets claude/pr-c-guardrail-redos-hardening (fix(signals): cap globToRegExp wildcard count to prevent ReDoS #2445) rather than main because it imports countWildcardGroups/MAX_GLOB_WILDCARD_GROUPS from that PR's change-guardrail.ts changes; rebasing onto main once fix(signals): cap globToRegExp wildcard count to prevent ReDoS #2445 merges is expected to be trivial (this PR only touches focus-manifest.ts and its test file beyond the shared import, neither touched by fix(signals): cap globToRegExp wildcard count to prevent ReDoS #2445 itself).
• Flagged as a follow-up during fix(signals): cap globToRegExp wildcard count to prevent ReDoS #2445's own adversarial review (the gate found the original ReDoS fix's threshold was inconsistent with its own benchmark data) — this closes the parallel gap in the sibling file that has the same class of counting bug but isn't a live exploit path today.

[loopover-orb]

Warning

🟨🟨🟨🟨🟨🟨🟨🟨🟨🟨🟨🟨

⏸️ Gittensory review result - manual review recommended

_{Review updated: 2026-07-02 01:45:41 UTC}

3 files · 1 AI reviewer · no blockers · readiness 86/100 · CI pending · unstable

⏸️ Suggested Action - Manual Review

Review summary
The change correctly moves focus-manifest glob validation from raw `*` character counting to the same wildcard-group accounting used by `globToRegExp`, so `**/*.json` no longer gets rejected while over-complex grouped globs are rejected before matching. Exporting the guardrail constant and counter from `src/signals/change-guardrail.ts` is coherent here because that module remains dependency-free, and the added regression test covers the previously rejected globstar-plus-star shape. The visible diff is focused and preserves the existing fail-closed parse-time behavior for unsafe content-lane globs.

Nits — 4 non-blocking

• nit: `test/unit/focus-manifest.test.ts:1096` still only proves rejection far above the cap with five groups; add a boundary regression for exactly three groups, e.g. `registry/*-*-*-final.json`, since that is the specific raw-count-equals-old-cap case this PR calls out.
• nit: `src/signals/focus-manifest.ts:623` has an unusually long explanatory comment with escaped example text; trimming it to the invariant and pointing to `change-guardrail.ts` would reduce future drift in prose while keeping the code-level source of truth intact.
• Add a focused test in `test/unit/focus-manifest.test.ts` that `contentLane.entryFileGlob: "registry/*-*-*-final.json"` is rejected with the warning, so the new group cap is locked at the dangerous boundary rather than only far beyond it.
• Shorten the `normalizeOptionalGlob` comment in `src/signals/focus-manifest.ts` to state that it reuses `countWildcardGroups`/`MAX_GLOB_WILDCARD_GROUPS`, leaving the detailed ReDoS rationale in `src/signals/change-guardrail.ts` where the implementation lives.

Signal	Result	Evidence
Code review	✅ No blockers	1 reviewer
Linked issue	⚠️ Missing	No linked issue or no-issue rationale found.
Related work	✅ No active overlap found	No same-issue or scoped active PR overlap found.
Change scope	✅ 20/20	Low review scope from cached public metadata (size label size:M; no linked issue context).
Validation posture	✅ 25/25	PR body includes validation/test evidence.
Contributor workload	✅ 10/10	Author activity: 65 registered-repo PR(s), 55 merged, 563 issue(s).
Contributor context	✅ Confirmed Gittensor contributor	JSONbored; Gittensor profile; 65 PR(s), 563 issue(s).
Gate result	✅ Passing	No configured blocker found.

Review context

• Author: JSONbored
• Role context: owner (maintainer lane)
• Public audience mode: oss maintainer
• Lane context: Repository registration is not available in the local Gittensory cache.
• Public profile languages: not available
• Official Gittensor activity: 65 PR(s), 563 issue(s).
• PR-specific overlap: none found.

Contributor next steps

• Treat this as maintainer-lane context rather than normal contributor-lane activity.
• Explain no-issue PR.
• Triage stale or unlinked PRs.
• No action.
• Link the issue being solved, or explicitly explain why this is a no-issue PR.

Signal definitions

• Related work = same linked issue, overlapping active PRs, or title/path similarity.
• Change scope = cached public metadata such as size labels, draft state, and review-burden hints.
• Validation posture = whether the PR provides enough public validation/test evidence for maintainer review.
• Contributor workload = public contributor activity and cleanup pressure, not a repo-wide quality failure.
• Contributor context = public GitHub/Gittensor identity context; non-Gittensor status is not a blocker.

_{🟩 Safe / merged · 🟦 Advisory · 🟨 Held for review · 🟥 Blocked / closed}

---

💰 Earn for open-source contributions like this. Gittensor lets GitHub contributors earn for the work they already do — register to start earning →.

Checked by Gittensory, a quiet PR intelligence layer for OSS maintainers.

• Re-run Gittensory review

[JSONbored]

Superseded by 536e46a ("fix(content-lane): share the wildcard-safety predicate between glob parsing and compilation"), already merged into claude/pr-b-content-lane-genericity — same fix, done concurrently in another session. Closing to avoid duplicate/conflicting work.

[codecov]

Codecov Report

✅ All modified and coverable lines are covered by tests.
⚠️ Please upload report for BASE (claude/pr-c-guardrail-redos-hardening@8ef2e04). Learn more about missing BASE report.
✅ All tests successful. No failed tests found.

Additional details and impacted files

@@                           Coverage Diff                            @@
##             claude/pr-c-guardrail-redos-hardening    #2448   +/-   ##
========================================================================
  Coverage                                         ?   95.92%           
========================================================================
  Files                                            ?      225           
  Lines                                            ?    25301           
  Branches                                         ?     9204           
========================================================================
  Hits                                             ?    24269           
  Misses                                           ?      419           
  Partials                                         ?      613           

Files with missing lines	Coverage Δ
src/signals/change-guardrail.ts	100.00% <100.00%> (ø)
src/signals/focus-manifest.ts	99.24% <100.00%> (ø)

🚀 New features to boost your workflow:

• 📦 JS Bundle Analysis: Save yourself from yourself by tracking and limiting bundle sizes in JS merges.