agentleFS
xialangji-debugxialangji-debug/codex-skills-agentmd-obsidian/skills/zentao-bug-resolver/SKILL.md

zentao-bug-resolver

Skill3 starsChanged 2 months agoMITView on GitHub

Resolve explicitly selected Zentao Bugs after firmware fixes or triage decisions are complete. Use for 标记已解决, 关禅道, 外部原因, resolve-plan review, or corrective reactivation of an accidentally resolved/closed Bug. Directly submit ordinary fixed Bugs when the current request already authorizes the exact IDs and current exact-product fix receipts exist; otherwise preview once.

The file

We split this skill into the rules it gives an agent

8 rules. About 0 are common, 4 are distinctive, and 4 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈4

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 4

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    Do not run preview and submit as two automatic rounds after the current request already satisfies the direct-submit row.

    line 19
  2. Distinctive:

    Call the resolver once:

    line 26
  3. Distinctive:

    Create a plan only when the selected Bugs require different values or a non-fixed decision.

    line 39
  4. Very specific:

    After the user approves that preview, rerun the same plan with --submit.

    line 46
  5. Very specific:

    Use fixed only for a current code-side fix or evidence that the issue is already fixed in the named build.

    line 49
  6. Very specific:

    Do not resolve ignored-items.md entries unless the user moves them into scope.

    line 66
  7. Distinctive:

    Use the latest activation note for a reactivated Bug.

    line 67
  8. Very specific:

    After a successful fixed submit, accept the script's exact-target zentao_resolved memory event.

    line 72

≈ “Common” is an estimate: at least 1 in 5 of the 350,495 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

Zentao Bug Resolver

This Skill owns development-side Zentao resolution only. 已解决 remains QA pending; never close a Bug for QA/test.

Choose One Path

Current request/evidencePath
Exact IDs explicitly authorized, every item is ordinary fixed, and each has a current exact-product fix receiptDirect minimal submit
User asks for a plan, authorization is absent, fields differ, any item is external/other, or evidence is incompletePreview once, then submit after approval
Accidental close, wrong-product resolution, or reactivationRead corrective operations

Do not run preview and submit as two automatic rounds after the current request already satisfies the direct-submit row.

Direct Minimal Submit

  1. Reuse the current fix receipt and task snapshot. Let the resolver check the live checkout and mapping required for submission.
  2. Confirm the exact Bug IDs and current fix receipts.
  3. Call the resolver once:
node "$env:USERPROFILE\.codex\skills\zentao-bug-resolver\scripts\zentao_bug_resolver.js" --repo . --ids 2799,2917 --minimal --submit

--minimal selects fixed, preserves the assignee, leaves the comment empty, and defaults the resolved build to trunk. Use --build-current-branch only when the requested resolved build is the current branch. The script still verifies product, current status, resolved build, form values, submission, and the final 已解决 detail page.

Preview Path

Create a plan only when the selected Bugs require different values or a non-fixed decision. Read resolve plan format, then run:

node "$env:USERPROFILE\.codex\skills\zentao-bug-resolver\scripts\zentao_bug_resolver.js" --repo . --plan ".codex\zentao-resolve-plan.md"

After the user approves that preview, rerun the same plan with --submit. Do not ask for a second approval already supplied in the current request.

Use fixed only for a current code-side fix or evidence that the issue is already fixed in the named build. Use external only when concrete evidence assigns the failure to platform/backend/data/account/network/applet/test environment; include a concise required remark. Do not infer either choice from a title.

Hard Boundaries

  • Resolve only explicitly selected IDs or IDs in an approved plan; never resolve the full fetched list.
  • Require detail-page 所属产品 to exactly match the confirmed 禅道产品 in the shared live ASR context and existing private project map. Missing, ambiguous, unconfirmed, or branch-mismatched mappings stop before browser access. Legacy zentao.md mappings are supported only for historical contexts outside a live repository. An explicit product cannot override a conflicting current mapping; prefix similarity is not enough.
  • --allow-product-mismatch is only for explicitly authorized corrective reactivation with --reactivate-resolved, never ordinary resolution.
  • Preserve the current assignee unless the user explicitly changes it.
  • Do not resolve ignored-items.md entries unless the user moves them into scope.
  • Use the latest activation note for a reactivated Bug.
  • Preserve unrelated local changes.
  • Load credentials only from the existing private Zentao credential source. Never print or store passwords, service URLs, or tokens in reports or memory.

After a successful fixed submit, accept the script's exact-target zentao_resolved memory event; it must not upgrade QA verification. Report IDs actually submitted, final statuses, failures, resolved-build fallback, and memory event count. An unmatched memory event stays unmatched; never link by title.

Resource

scripts/zentao_bug_resolver.js performs all product/status/build/form/final-page checks and writes the operation report. Do not duplicate those checks outside the script.

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.