AgentStack
Browse Sign in
Browse Why AgentStack Sell Docs
Sign in
SKILL verified MIT Self-run

Task Metadata Audit

skill-choral-io-choral-skills-task-metadata-audit · by choral-io

Use when task items or Kanban links need a read-only check for readiness, dependencies, issues, acceptance, or board consistency.

No reviews yet
0 installs
6 views
0.0% view→install

Install

$ agentstack add skill-choral-io-choral-skills-task-metadata-audit

✓ scanned · ✓ verified, works with Claude Code, Cursor, and more.

Security review

✓ Passed

No issues found. Passed automated security review. · v0.1.0 How review works →

  • Prompt-injection patterns
  • Secret / credential exfiltration
  • Dangerous shell & filesystem operations
  • Untrusted network calls
  • Known-malicious package signatures

What it can access

  • Network access No
  • Filesystem access No
  • Shell / process execution No
  • Environment & secrets No
  • Dynamic code execution No

From automated source analysis of v0.1.0. “Used” means the capability is present in the source — more access means more to trust, not that it’s unsafe.

View the full security report →

Verified badge

Passed review? Show it. Paste this badge into your README, it links to the public security report.

AgentStack Verified badge Links to your public security report.
[![AgentStack Verified](https://agentstack.voostack.com/badges/verified.svg)](https://agentstack.voostack.com/security/report/skill-choral-io-choral-skills-task-metadata-audit)

Reliability & compatibility

Security review passed
0 installs to date
no reviews yet
1mo ago

Declared compatibility

Claude CodeClaude Desktop

Compatibility is declared by the source manifest. End-to-end runtime verification is coming, see below.

Preview Execution monitoring

We're building live execution health for every listing: tool-call success rate, median latency, uptime, and last-checked timestamps, measured, not self-reported. It isn't live yet, so we don't show numbers we can't stand behind.

How agent discovery & health will work →
Are you the author of Task Metadata Audit? Claim this listing to set pricing, connect Stripe payouts, and keep 70% of every sale.
Sign up to claim

About

Task Metadata Audit

Runtime Context

Before acting, resolve ` using the runtime bootstrap rules, then read /.workflow/runtime.md and /.workflow/manifest.yml`; do not assume non-default workflow paths or default ids.

Use this skill to inspect task metadata and report issues. This skill is read-only.

Workflow

  1. Read /.workflow/schemas/common.md.
  2. Read /.workflow/schemas/tasks.md.
  3. Read /.workflow/rules/delivery.md.
  4. Read /planning/KANBAN.md.
  5. Scan /tasks/*.md, excluding localized files.
  6. Parse frontmatter and task body sections.
  7. Check task relationships against existing task ids and Kanban card ids.
  8. Report issues and dry-run fixes without editing files.

Checks

  • Missing or invalid type, priority, value, effort, readiness, or module.
  • Missing or invalid severity only when type is issue, bug, or defect and impact is needed for triage.
  • Missing, empty, or unresolved owners in task items that are readiness: ready, linked from a Kanban card, scheduled, assigned, or actively maintained.
  • owners, assignees, or reviewers values that reference member or group documents without wikilinks, or tool-written values that should be path-qualified.
  • Group values in assignees; report these as team-pool assignments, not direct assignment to the current member.
  • Missing assignees only when the task appears scheduled or actively assigned.
  • Missing reviewers only when review responsibility is expected.
  • blocked_by or related_to entries that point to missing or ambiguous task references.
  • Missing ## Sources section or sources that point only to local workspace notes or localized files.
  • Tasks with readiness: ready but missing acceptance criteria.
  • Issue, bug, or defect tasks with readiness: ready but missing Problem, Impact, or Triage context.
  • Reproducible bug or defect tasks with readiness: ready but missing Reproduction, Expected, or Actual.
  • Tasks with readiness: ready whose task item or required source materials are uncommitted.
  • Tasks with readiness: ready whose task item or required source materials are not pushed to the default remote when one exists.
  • Tasks in Ready with unresolved blocked_by entries.
  • Tasks in Ready with readiness other than ready.
  • Tasks in Blocked without readiness: blocked, unresolved blocked_by, or an explicit blocker note.
  • Tasks in Backlog with readiness: blocked; report as allowed planned dependency state, not an error.
  • Tasks outside Backlog or Blocked with unresolved blocked_by entries.
  • Downstream tasks whose blocked_by references are all resolved but still have readiness: blocked or remain in Blocked.
  • Tasks in Backlog that appear ready but have not been promoted to Ready.
  • Localized task files used as planning inputs.

Output

Use the standard task audit output shape from references/report.md: Summary, Findings, Dry-run fixes, Requires judgment, Safe for next-task-selection, and Sources.

Guardrails

  • Do not edit task items.
  • Do not move Kanban cards.
  • Do not infer sensitive information.
  • Do not treat assignees as Kanban status.
  • Treat source stability as a readiness audit and Ready-column maintenance check. Do not require next-task-selection to run the full audit for every candidate unless the user asks for strict verification or automatic start.
  • Require maintainer approval before any suggested fix is applied by another skill.

References

  • For audit output examples, read references/report.md.

Source & license

This open-source skill is cataloged on AgentStack and links to its original source — we do not rehost the code.

Install and usage instructions live in the source repository linked above.

Reviews

No reviews yet, be the first.

Versions

  • v0.1.0 Imported from the upstream source.