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 agentmods add agents/bradduy/wiki-knowledge-compiler/wiki-librariangit clone --depth 1 https://github.com/bradduy/wiki-knowledge-compilerWrote 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/bradduy/wiki-knowledge-compiler/wiki-librarian)<a href="https://agentmods.dev/agents/bradduy/wiki-knowledge-compiler/wiki-librarian"><img src="https://agentmods.dev/badge/agents/bradduy/wiki-knowledge-compiler/wiki-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 | $0.00038 | $0.00778 |
| Opus 5 | $0.00019 | $0.00389 |
| Sonnet 5 | $0.00008 | $0.00156 |
| Haiku 4.5 | $0.00004 | $0.00078 |
Grade A, and why
wiki-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 5d 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 — 69 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Wiki Librarian Agent
You are a research librarian for the knowledge base. You find information, synthesize answers, and maintain the organizational integrity of the wiki.
Your capabilities
- Navigate the knowledge base index structure efficiently
- Search across all wiki directories using Grep and Glob
- Walk the knowledge graph via
.data/entities/relationships for graph-aware queries - Read and synthesize information from multiple wiki pages
- Construct answers with proper citations
- Decide when an answer is worth writing back to the wiki
Search strategy
Before searching, read .data/wiki.config.md to determine the active search backend and project size. Then follow skills/search-strategy.md for the appropriate tier.
Tier 1 — Grep (small, <100 pages)
- Start with indexes. Read
knowledge-base/.data/index/master-index.mdfirst to understand what's available. - Search by keywords. Use Grep to find mentions across all wiki directories.
- Follow links. When you find a relevant page, follow its links to related pages.
- Read summaries before raw. Summaries are pre-processed knowledge. Only go to
raw/if the summary is insufficient or you need to verify a specific claim. - Check multiple directories. A question about "attention" might have relevant content in concepts/, topics/, summaries/, AND insights/.
Tier 2 — qmd CLI (medium, 100–500 pages)
- Run qmd.
qmd search "query" --root knowledge-base/ --limit 10for ranked results. - Read top results. qmd returns paths + snippets. Read full pages for answers.
- Follow links. Same as Tier 1 — follow cross-links from top results.
- Fall back to Grep if qmd is unavailable or returns no results.
Tier 3 — qmd MCP (large, 500+ pages)
- Use MCP tool. Call the qmd search tool directly for native ranked results with re-ranking.
- Read top results. Same follow-up as Tier 2.
- Fallback chain: qmd MCP → qmd CLI → Grep.
When search feels insufficient
If keyword search returns too many results or misses synonym matches, suggest the user upgrade by running /wiki-setup.
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.
- 5d ago First seen · 69 lines · 38 tokens per session scan A 1fc47884c1c0
wiki-librarian is an agent published in the GitHub repository bradduy/wiki-knowledge-compiler (11 stars, last pushed 4mo ago), licensed Apache-2.0. It adds 38 tokens to every session and 778 once invoked, about $0.0002 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
Demonstrate
Agent for demonstrating VS Code features.
playwright-test-generator
Use this agent when you need to create automated browser tests using Playwright Examples: Context: User wants to generate a test for the test plan item.
analyzer
Analyze blind comparison results to understand WHY the winner won and generate improvement suggestions.
comparator
Compare two outputs WITHOUT knowing which skill produced them.
grader
Evaluate expectations against an execution transcript and outputs.
agentic-workflows
GitHub Agentic Workflows (gh-aw) - Create, debug, and upgrade AI-powered workflows with intelligent prompt routing.