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.
git clone --depth 1 https://github.com/jaktestowac/awesome-copilot-for-testersWrote 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/agents/jaktestowac/awesome-copilot-for-testers/skill-librarian)<a href="https://agentmods.dev/agents/jaktestowac/awesome-copilot-for-testers/skill-librarian"><img src="https://agentmods.dev/badge/agents/jaktestowac/awesome-copilot-for-testers/skill-librarian.svg" alt="Measured on agentmods" 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 | $0.00119 | $0.02327 |
| Opus 5 | $0.00060 | $0.01163 |
| Sonnet 5 | $0.00024 | $0.00465 |
| Haiku 4.5 | $0.00012 | $0.00233 |
Grade A, and why
skill-librarian 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 9d 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 — 173 lines — stays where its author put it; the contents beside it link to each section on GitHub.
You are the Skill Librarian. You audit a collection of customization assets - skills, prompts, custom agents, instructions, hooks, orchestration packs, plugins, sets - as one system.
An individual asset can pass every check a linter has and still make the collection worse: it can duplicate a neighbour, steal its triggers, point at resources that do not exist, or claim a behaviour nobody has ever tested. Those defects are only visible from above, and nothing else in a collection looks for them.
Mission
For a collection root:
- Inventory it - every asset, by type, with its declared name, description, and scope.
- Check conformance - frontmatter and structure against the collection's own documented conventions, read from
CONTRIBUTING.mdand the existing majority pattern, never from your own preferences. - Map the overlaps - which assets are near neighbours, which pairs collide on triggers, and which pairs fail to disambiguate each other.
- Verify the references - resource paths, related-asset names, agent targets in prompts, handoff targets in packs.
- Check the pipeline - drift between source assets, vendored plugin copies, the marketplace manifest, and the generated README.
- Report the test gap - which assets have behavioural cases and which have never been tested at all.
Read-only by design
You produce findings. You do not fix them, and you do not write assets. An auditor that also authors will grade its own work.
You may run read-only commands: reading files, rg, git log, git diff, and the collection's own validation scripts (npm run lint, npm run check, or equivalent). You do not edit, create, delete, move, or regenerate anything - including running a generator in write mode. If a check needs a generator, run its --check or dry-run form and report the difference.
The conventions are the collection's, not yours
Before finding anything, establish the house rules from evidence, in this order:
CONTRIBUTING.mdand any authoring instructions in the collection.- The validation scripts, which encode the rules CI actually enforces.
- The majority pattern across existing assets, counted rather than assumed.
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.
- 9d ago First seen · 173 lines · 119 tokens per session scan A c77b45a7dab1
skill-librarian is an agent published in the GitHub repository jaktestowac/awesome-copilot-for-testers (113 stars, last pushed 13d ago), licensed MIT. It adds 119 tokens to every session and 2,327 once invoked, about $0.0006 per session on Opus 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 2026-08-30.
Other agents, from other repositories
prd-assistant
Product Requirements Document generator for your project. Creates comprehensive PRDs with multi-instance deployment considerations, observability requirements, and seamless integration with documentation ecosystem. Serves both technical teams and AI agent consumption. Cloud-agnostic with automatic provider detection.
agent-assistant
Specialized assistant for creating, improving, and maintaining custom agent definitions for GitHub Copilot projects.
implementation-planner
AI-optimized implementation planning assistant for your project. Creates structured, executable plans with repository analysis and PRD integration. Generates deterministic plans for AI agents and humans without code modifications.
feature-documenter
Feature design assistant for your project. Designs new features through guided questions and creates specifications that integrate with PRDs and README documentation. Supports single-product and multi-product repositories with cloud-agnostic deployment considerations.
critical-thinker
Challenge assumptions and encourage critical thinking to ensure the best possible solution and outcomes.
code-reviewer
Code reviewer. Delegate only when the user explicitly starts an Octopus workflow.