Getting it into your agent
This one installs as part of its plugin. Adding the marketplace and installing the plugin brings it with everything else the plugin ships.
/plugin marketplace add kubalajiri/claude-squirrel/plugin install squirrelWrote 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/kubalajiri/claude-squirrel/squirrel-rules)<a href="https://agentmods.dev/commands/kubalajiri/claude-squirrel/squirrel-rules"><img src="https://agentmods.dev/badge/commands/kubalajiri/claude-squirrel/squirrel-rules.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 | $0.00030 | $0.00630 |
| Opus 5 | $0.00015 | $0.00315 |
| Sonnet 5 | $0.00006 | $0.00126 |
| Haiku 4.5 | $0.00003 | $0.00063 |
Grade A, and why
squirrel-rules scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 6d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
What it actually says
Run this exact command with the Bash tool and show the user its output verbatim:
sh "${CLAUDE_PLUGIN_ROOT}/scripts/squirrel.sh" --rules "$ARGUMENTS"
With no arguments this lists every rule in evaluation order — the shipped idle ladder first,
then the user's own — marking the one that is winning right now (<== WINNING NOW), along
with the metrics available and any rule that does not parse. That listing is the answer to
"why is my bar this colour?"; read it out rather than reasoning about precedence.
A condition is exactly <metric> <op> <value>, or a bare <metric> for true/false:
weekly.pct >= 80 git.dirty cost.usd > 5 git.branch == main
Operators are >= <= == != > <. Run the command with no arguments to get the live
list of metrics — they cover context and usage percentages, cost, duration, lines, tokens,
git (dirty/ahead/branch), repo, model, effort, idle seconds, and activity state.
Nothing here is ever evaluated as code. A condition that does not parse — an unknown metric,
a missing value, two comparisons — is reported and ignored, and the bar keeps rendering.
/squirrel-rules and /squirrel-config both say how many rules are being ignored and why.
A metric with no value never matches, not even with !=. "Unknown" is not "different
from": a rule must not paint the bar because data is missing.
add <condition> bg=.. text=.. [bold=on|off] [priority=<n>]— a rule must paint something, sobg,text, or both are required.set <condition> <field>=<value> ...— overlays only the fields given onto an existing rule.remove <condition>— drops one of the user's own rules (the idle ladder is edited with/squirrel-phases, not here).reset— clears the user's rules; the idle ladder remains.
Which rule wins: the last matching rule, unless one carries a higher priority. The idle
phase ladder ships as the first rules, so a user rule added later wins a tie with it by
default — mention that if they ask why their rule beats the idle tint.
Every edit validates before writing; a refused edit changes nothing. Never write the
barRules config key by hand — always go through this command.
Do not do anything else.
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.
- 6d ago First seen · 45 lines · 0 tokens per session scan A f821af72b7b1
squirrel-rules is a command published in the GitHub repository kubalajiri/claude-squirrel (2 stars, last pushed 11d ago), licensed MIT. It adds 30 tokens to every session and 630 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 2026-08-31.
Other commands, from other repositories
todo
The quality-gated task list: tasks with real descriptions, testable acceptance criteria, and evidence — a task only closes when the controller agrees it is done.
ask
Ask a single expert agent a question - like a Slack ping to a colleague.
rows
Configure which rows cc-statusline shows (or turn it off entirely).
config
Configure iterm2-tab-status plugin settings (emoji prefixes, colors, display target, subtitle activity, badge, notifications).
statusline
Show the governor fleet in your Claude Code statusline. Explicit, opt-in, and chains — it never replaces an existing statusline.
pin
Break checkpoint — status display, session log, optional MEMORY.md update.