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/commands/coltonbearden/carrel/search)<a href="https://agentmods.dev/commands/coltonbearden/carrel/search"><img src="https://agentmods.dev/badge/commands/coltonbearden/carrel/search/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/commands/coltonbearden/carrel/search"><img src="https://agentmods.dev/badge/commands/coltonbearden/carrel/search.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 | $0.00029 | $0.00789 |
| Opus 5.5 | $0.00012 | $0.00316 |
| Sonnet 5.5 | $0.00006 | $0.00158 |
| Haiku 4.5 | $0.00003 | $0.00079 |
Grade A, and why
search 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.
What it actually says
Search the desk index for: $ARGUMENTS
Run the carrel CLI via Bash. Map the user's request onto the real flags in the --help block below (regenerated from the CLI by scripts/sync_plugins.py; if the installed carrel search --help differs, trust the installed version — never invent flags):
Usage: carrel search [OPTIONS] QUERY
Full-text search the desk index for QUERY (FTS5 syntax, bm25-ranked).
Matched terms are bracketed in the snippet. Filters combine with AND. JSON output is a list of
{"path", "score", "snippet"} (lower bm25 score = better match). Run the `index` command first to
build the index under --root.
Options:
--limit INTEGER Maximum number of hits. [default: 20]
--type T1,T2 Only these file types, comma-separated (e.g. pdf,md).
--tag TAG Only files carrying TAG (repeatable — every TAG must match).
--meta CONDITION Only files whose fields satisfy CONDITION, e.g. vendor=acme, total>1000,
due<2026-11, paid? (repeatable — every one must hold; see `meta find`).
--fail-empty Exit 5 when there are no hits.
--json Machine-readable JSON output.
--help Show this message and exit.
Note: --json and --root are global flags and go before search (carrel --json --root DIR search "QUERY").
QUERYuses FTS5 syntax: quoted phrases ("exact phrase"),AND/OR/NOT, prefix matching (term*). Translate natural-language requests into a sensible FTS5 query.--type T1,T2: restrict to file types (e.g.pdf,md) — use it when the user says "which PDFs..." or "in my notes".--tag TAG(repeatable, ANDed): only files carrying every given tag — pairs with/carrel-organize:tag.--limit N(default 20).--fail-empty: exit 5 instead of an empty list when nothing matches (handy in scripts).--root DIR: use it when the user's desk lives somewhere other than the cwd.
The index must exist first: if the search errors because there is no index under --root, run carrel index there (add --ocr only if the user wants image/scan text searchable) and retry, telling the user what you did.
Always use --json and interpret the {path, score, snippet} hits (lower bm25 score = better): present the top hits with their snippets, matched terms are bracketed. No hits → suggest a broader query, a different --type, or re-indexing. To turn the hits into context for a longer answer, hand the same query to /carrel-inspect:pack --query.
Requires the carrel CLI on PATH. If carrel is not found, tell the user to install it with uv tool install carrel (see the repo's INSTALL notes), or run it as uv run carrel ... from the carrel repo root.
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 · +2 lines 3f88ecfcb050
- 1mo ago Changed · +17 lines · +6 tokens per session 9992536a5a5a
- 1mo ago Changed 3ae3cfcc842d
- 1mo ago First seen · 28 lines · 23 tokens per session scan A 57819fb009f6
search is a command published in the GitHub repository coltonbearden/carrel (2 stars, last pushed yesterday), licensed MIT. It adds 29 tokens to every session and 789 once invoked, about $0.0001 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 commands, from other repositories
add-agent
Add a new sub-agent to the current plugin for specialized tasks.
add-command
Add a new slash command to the current plugin.
add-skill
Add a new Skill to the current plugin with proper structure.
init
Scaffold a new Claude Code plugin with standard structure and starter files.
hello
Greet the user with a friendly, personalized message.
run-quality-gates
Run code quality checks (lint, format, type check, tests).