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 commands/mozilla-ai/cq/statusgit clone --depth 1 https://github.com/mozilla-ai/cqWhat 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 | $0.00027 | $0.00612 |
| Opus 5 | $0.00014 | $0.00306 |
| Sonnet 5 | $0.00005 | $0.00122 |
| Haiku 4.5 | $0.00003 | $0.00061 |
Grade A, and why
cq:status 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 yesterday.
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 — 61 lines — stays where its author put it; the contents beside it link to each section on GitHub.
/cq:status
Display a summary of the cq knowledge store.
Instructions
- Call the
statusMCP tool (no arguments needed). - Format the response as a readable summary using the sections below.
Output Format
Present the results using this structure:
## cq Knowledge Store
### Tier Counts
local: {count} | private: {count} | public: {count}
*local* = on this machine only; *private* = shared with everyone who can reach the same `CQ_ADDR`; *public* = the open commons, shared across every node that participates in it (available when your configured remote serves it).
### Recent Local Additions
- {id}: "{summary}" ({relative time})
- ...
### Confidence Distribution
*(covers local plus your private/org units; excludes the public commons)*
■ 0.7-1.0: {count} units
■ 0.5-0.7: {count} units
■ 0.3-0.5: {count} units
■ 0.0-0.3: {count} units
### Domains
{distinct count} total
{domain} ({count}), {domain} ({count}) ... +{remaining} more
The tier_counts field contains the tier breakdown. Display all tiers present in the response. Omit tiers with a count of 0.
The recent field reflects the local store only. When a remote is configured and reachable, units proposed via Client.Propose go directly to the remote and do not appear here. If recent is empty, render the section as (no recent local additions) so users understand the scope.
The domain_counts field maps each domain to its unit count. Domains are the least important section, so render them last and keep them compact: a {distinct count} total line, then the most-tagged few on the next line as {domain} ({count}) (highest count first, ties alphabetical), capped at 8. If more than 8 exist, append the truncation marker after a single space, e.g. {domain} ({count}) ... +{remaining} more; do not leave a trailing comma before it.
If the warnings field is non-empty, surface each entry prominently above the summary so the counts are not mistaken for the full picture. A warning means stats aggregation degraded; for example, an unreachable or misconfigured remote, in which case the counts reflect the local store only:
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.
- yesterday First seen · 61 lines · 27 tokens per session scan A 8b31b3af5beb
cq:status is a command published in the GitHub repository mozilla-ai/cq (1,258 stars, last pushed 17d ago), licensed Apache-2.0. It adds 27 tokens to every session and 612 once invoked, about $0.0001 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 commands, from other repositories
api-update
Check for open PRs in the Composio client repositories and create a plan for updating the SDKs.
station
You are helping the user work with Station - the self-hosted AI agent orchestration platform.
add-tool
Add a new MCP tool to server.py.
add-adapter-method
Add a new method to both Airflow adapters.
check-airflow-compat
Verify code works with both Airflow 2.x and 3.x.
checklist
Generate a custom checklist for the current feature based on user requirements.