task_audit SKILL.md skillB
task_audit from theafh/ai-modules is an agent skill: a SKILL.md folder an AI agent loads when a task calls for it.
mdr add theafh/ai-modules/task_audit@v1.0.16Pins v1.0.16 in mdr.lock. Pin the fingerprint instead.
- Audit grade
- B, 16 of 17 checks passed
- Kind
- SKILL.md skill
- Repository
- theafh/ai-modules
- Path
plugins/ai_dev/skills/task_audit/SKILL.md- Versions recorded
- 7
- Latest version
- v1.0.16, committed Oct 13, 2026
- First version recorded
- Jul 20, 2026
- Size
- 9,127 bytes
- Repository stars
- 39
- License
- MIT
Audit one implemented or finished task against the actual codebase. Use when verifying a believed-done task, checking claimed completion against the code, or running a drift check before close-out. Inspect code and tests, run verification, stamp audited only on clean implemented work, and report gaps otherwise.
Collected from public GitHub. Every version is saved exactly as it was and never changes. Install it at an exact version with the mdr tool, and every file is checked before it reaches your agent. The main file is checked against the SHA-256 fingerprint recorded here, and the others against the fingerprints GitHub keeps for its source commit. The automatic checks below grade the latest version, and the same checks always give the same file the same grade.
What the file says
# task_audit <task_audit_skill> <role> task_audit verifies whether a task's *claimed* completion is *real*, by checking the codebase rather than the prose. It reads one task, reads the actual implemented code and tests, walks every body item and `## Acceptance` check, runs the suite the task names, and reports a verdict: clean, or a list of gaps with fixes. It is a gate with two allowed lifecycle mutations: on a clean, complete verdict over a current `implemented` task, it stamps `status: audited` and bumps `updated`, and on a gap verdict over an `audited` task, it re-opens the task to `status: implemented` and bumps `updated`, the base `<status_discipline>` re-audit re-open. A clean re-check of an archived `finished` task leaves it `finished`, and a gap on an `implemented` or archived task leaves the status as it is. It answers "is this task genuinely done?"; `task_finish` answers "now close it." On a clean pass it hands the close-out to `task_finish`; on gaps it hands the remaining work on an `implemented` or re-opened task to `task_implement`, and the gaps on an archived task to a new task under the base `<status_discipline>` follow-up rule. </role> <when_to_activate> …
Read the whole file at its exact version.
How to install
mdr add theafh/ai-modules/task_audit@sha256:f2a162d5beeeb00fThe line at the top of this page pins the label v1.0.16, which is what most people want. This one pins the sha256 fingerprint of the contents instead, so it survives a label being moved. Either way the exact version is written to mdr.lock, and mdr install downloads and checks those contents again, so it installs them exactly or stops with an error.
[](https://markdownregistry.com/a/art_syzn225yq56he5ni)
3 badge image fetches in 30 days. What the badge shows.
Versions
The registry has recorded 7 versions of this file, between Jul 20, 2026 and Oct 13, 2026, about one every 14 days.
| version | committed | commit | size | audit | |
|---|---|---|---|---|---|
| v1.0.16 latest | 2026-10-13 | 0cf0227 | 9,127 B | B | view · diff |
| v1.0.15 | 2026-10-11 | b011c11 | 8,648 B | B | view · diff |
| v1.0.14 | 2026-09-05 | e04ae6b | 8,193 B | B | view · diff |
| v1.0.13 | 2026-08-11 | 2a65443 | 8,252 B | B | view · diff |
| v1.0.12 | 2026-08-06 | a895d2f | 8,185 B | B | view · diff |
| v1.0.11 | 2026-07-21 | 30b7173 | 7,663 B | B | view · diff |
| v1.0.10 | 2026-07-20 | 416a546 | 7,606 B | B | view |
Audit of the latest version
- fail: No script tag (matched: <script)
- pass: Frontmatter block present
- pass: Frontmatter declares a name
- pass: Frontmatter declares a description
- pass: Size between 200 bytes and 200 KB (9127 bytes)
- pass: No zero-width or bidi control characters
- pass: No instruction hidden inside an HTML comment
- pass: No link to an exfiltration or paste host
- pass: No credential-shaped string
- pass: No instruction to send local credentials anywhere
- pass: No text hidden with inline styles
- pass: No prompt-injection phrasing
- pass: No curl or wget piped into a shell
- pass: No recursive delete of root, home or parent
- pass: No instruction to read or print local credentials
- pass: No base64 blob over 200 characters
- pass: No link to a raw IP address
Starting a new one? Use a template for SKILL.md on mdpantry, which passes every one of these checks and says what to change in it.
Source
theafh/ai-modules · 39 stars · license MIT · pushed 2026-11-08 · branch main
API
GET https://markdownregistry.com/api/v1/artifacts/art_syzn225yq56he5ni GET https://markdownregistry.com/api/v1/resolve?ref=theafh/ai-modules/task_audit GET https://markdownregistry.com/api/v1/blob/f2a162d5beeeb00fd5807ff0bfe7e5ead54e69b5fa6241dc32752931e5006083
Your AI assistant does the legwork. You hear about the deals worth your word. Hand yours the standing instructions at modelranch.com and it joins the network that reads files like this one.
Guides: the SKILL.md format and its required fields · what to check before you install a skill · how to pin a skill to an exact version · where to find agent skills · keeping a team on the same skills · which agent file does what.