Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx skills add bartoszarendt/agenticloop --skill work-unit-auditgit clone --depth 1 https://github.com/bartoszarendt/agenticloopWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/skills/bartoszarendt/agenticloop/work-unit-audit)<a href="https://agentmods.dev/skills/bartoszarendt/agenticloop/work-unit-audit"><img src="https://agentmods.dev/badge/skills/bartoszarendt/agenticloop/work-unit-audit/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/skills/bartoszarendt/agenticloop/work-unit-audit"><img src="https://agentmods.dev/badge/skills/bartoszarendt/agenticloop/work-unit-audit.svg" alt="Reviewed on agentmods" width="80" height="20"></a>What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00083 | $0.03777 |
| Opus 5.5 | $0.00033 | $0.01511 |
| Sonnet 5.5 | $0.00017 | $0.00755 |
| Haiku 4.5 | $0.00008 | $0.00378 |
Grade A, and why
work-unit-audit scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 1mo ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 404 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Work-unit audit
Task review proves each task was done correctly. Work-unit audit answers the question no task review asks: does the combined result achieve the work unit's intended outcome as a complete, coherent, sufficiently verified system?
Individually accepted tasks can still leave missing cross-task behavior, an incomplete product outcome, conflicting configuration, duplicate mechanisms, code and docs that disagree, unnecessary combined complexity, integrated regressions, or task-level evidence that never proved the final combined state.
Work-unit audit is enabled by default. Unless a human explicitly records
work_unit_audit: disabled in .agenticloop/project.md, work-unit closeout
cannot publish AGENT_CLOSEOUT_STATUS: complete without a current certificate.
0. Entry guard
This skill governs Agentic Loop-mode work-unit certification. It applies when the Auditor is invoked in Agentic Loop mode to certify or re-audit a tracked work unit against a designated audit record or audit packet.
It does not automatically apply to a standalone Auditor assessment. A standalone auditor takes its scope from the parent request, needs no audit packet, audit record, covered-task set, or frozen candidate, and returns non-certifying findings; it may consult this file as a reference without adopting the procedure. A standalone assessment never satisfies the certification gate below.
1. Preconditions
Do not start an audit until all of the following hold:
- every covered task is
acceptedorclosedunder the configured backend; - the covered artifacts are integrated or composed into one exact candidate;
- the candidate is frozen: a commit SHA or another immutable revision reference;
- final integrated verification evidence exists for that exact candidate;
- an Auditor model is configured at
adapters.<host>.roleSettings.auditor.
If auditing is enabled and no Auditor model is configured, that is a blocking setup condition. Report it and stop. Never substitute the Maintainer model.
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 1mo ago Changed 5f4e0f57fe3c
- 1mo ago First seen · 404 lines · 83 tokens per session scan A ac2617ba25ee
work-unit-audit is a skill published in the GitHub repository bartoszarendt/agenticloop (2 stars, last pushed 1mo ago), licensed MIT. It adds 83 tokens to every session and 3,777 once invoked, about $0.0003 per session on Opus 5.5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.
Other skills, from other repositories
assignment-review
A review skill for checking submitted Python, Go, code explanations, debugging notes, or practice exercises.
code-review
Comprehensive code review for security, performance, and best practices.
ppt-generation
Use this skill when the user requests to generate, create, or make presentations (PPT/PPTX). Creates visually rich slides by generating images for each slide and composing them into a PowerPoint file.
image-generation
Use this skill when the user requests to generate, create, imagine, or visualize images including characters, scenes, products, or any visual content. Supports structured prompts and reference images for guided generation.
newsletter-generation
Use this skill when the user requests to generate, create, write, or draft a newsletter, email digest, weekly roundup, industry briefing, or curated content summary. Supports topic-based research, content curation from multiple sources, and professional formatting for email or web distribution. Trigger on requests…
data-analysis
Use this skill when the user uploads Excel (.xlsx/.xls) or CSV files and wants to perform data analysis, generate statistics, create summaries, pivot tables, SQL queries, or any form of structured data exploration. Supports multi-sheet Excel workbooks, aggregation, filtering, joins, and exporting results to…