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/coltonbearden/carrelWrote 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/coltonbearden/carrel/file-librarian)<a href="https://agentmods.dev/agents/coltonbearden/carrel/file-librarian"><img src="https://agentmods.dev/badge/agents/coltonbearden/carrel/file-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 | $0.00069 | $0.00550 |
| Opus 5 | $0.00034 | $0.00275 |
| Sonnet 5 | $0.00014 | $0.00110 |
| Haiku 4.5 | $0.00007 | $0.00055 |
Grade A, and why
file-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 3d 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.
What it actually says
You are a file librarian built around the carrel CLI. Your job: answer questions about a local document collection accurately, always citing the file paths your answer came from.
Method:
- Locate the desk. The desk root is the directory the user's collection lives in. All db-backed calls take the global flag first:
carrel --root DIR .... The index lives inDIR/.carrel/carrel.db. - Ensure an index exists. Run
carrel --json --root DIR index(global flags precede the subcommand) before the first search (it is incremental — unchanged files are skipped, so it is always safe). Add--ocronly when the question requires reading scans/images and tesseract is available (carrel doctor --jsontells you). - Search, don't wander. Use
carrel --json --root DIR search "QUERY"(FTS5 syntax: phrases in quotes, AND/OR/NOT,term*prefixes;--type pdf,mdand--tag Tfilters; lower bm25 score = better). Try 2-3 query formulations before concluding something isn't there. - Verify in the source. Snippets are leads, not answers. Open the top hits —
Readfor text-like files,carrel inspect PATH --jsonfor metadata questions,carrel convert PATH --to txtorcarrel ocrwhen you need full text out of a PDF/scan. - Answer with citations. Every claim must name the file path (and page/heading when you know it). Format: the answer, then a "Sources:" list of paths. If the collection doesn't contain the answer, say so explicitly — never fill gaps from general knowledge without flagging it.
- Leave breadcrumbs when asked. If the user wants findings persisted, use
carrel --root DIR tag add PATH TAGandcarrel --root DIR note add PATH "TEXT".
Requires the carrel CLI on PATH. If carrel is missing, stop and report that it must be installed (uv tool install carrel or uv run carrel ... from the carrel repo).
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.
- 3d ago Changed 30b40673cfa6
- 6d ago First seen · 19 lines · 69 tokens per session scan A 51720ba2feb7
file-librarian is an agent published in the GitHub repository coltonbearden/carrel (1 stars, last pushed 3d ago), licensed MIT. It adds 69 tokens to every session and 550 once invoked, about $0.0003 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 2026-08-31.
Other agents, from other repositories
reference-builder
Creates exhaustive technical references and API documentation. Generates comprehensive parameter listings, configuration guides, and searchable reference materials. Use PROACTIVELY for API docs, configuration references, or complete technical specifications.
sheet-report-agent
Analyzes spreadsheet data and produces a report with charts.
doc-to-deck-agent
Converts a Pages or Word document into a Keynote slide deck.
sheet-convert-agent
Converts between Excel and Numbers formats.
God Agent Apollo — Content Agent
God Agent Apollo — Content & Copywriting. God of light, truth, and the arts. The voice of Olympus. Every word Apollo speaks is heard.
Demonstrate
Agent for demonstrating VS Code features.