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/skills/kubalajiri/claude-squirrel/squirrel)<a href="https://agentmods.dev/skills/kubalajiri/claude-squirrel/squirrel"><img src="https://agentmods.dev/badge/skills/kubalajiri/claude-squirrel/squirrel/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/skills/kubalajiri/claude-squirrel/squirrel"><img src="https://agentmods.dev/badge/skills/kubalajiri/claude-squirrel/squirrel.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 | $0.00158 | $0.14351 |
| Opus 5 | $0.00079 | $0.07175 |
| Sonnet 5 | $0.00032 | $0.02870 |
| Haiku 4.5 | $0.00016 | $0.01435 |
Grade A, and why
squirrel scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 8d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 662 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Squirrel status line
Squirrel is a two-line Claude Code status line: context usage, 5-hour and weekly rate limits, git state, cost, and an activity chip that tells you when Claude is working, idle, or needs you. Everything about it is configured through one bundled script — never by hand-editing the user's files.
1. How to act
Always change settings by running the bundled script, never by hand-editing the user's
config file or settings.json:
sh "${CLAUDE_PLUGIN_ROOT}/scripts/squirrel.sh" <subcommand> <args>
- Show the command's own output back to the user verbatim — it is written for them.
- Run
--show-configfirst when the request depends on current values (e.g. "what's my alarm set to?", "make it a bit more sensitive"). - Run
--help-textif you are unsure of an exact key or subcommand name; it lists every subcommand, segment key, and tunable with its bounds, pulled live from the code so it can never drift out of date. - Every setting has a subcommand — see §4. Never hand-edit the config file; there is no setting Squirrel can't reach through a command.
2. What the bar shows
Line 1 — usage:
| Segment key | Shows |
|---|---|
context |
ctx ▰▰▰▱▱ 42% (84k/200k) — context-window used, as a bar and percentage |
tokens |
the (used/window) token-count suffix on the context segment |
session |
the 5-hour rate-limit bucket, as a bar and percentage — 19pp/90% when this session's own contribution to the account's usage is known ('19 percentage points of the account's 90%'), plain 90% otherwise; see §6 |
weekly |
the all-models weekly rate-limit bucket |
permodel |
per-model weekly buckets alongside the all-models one (needs weekly on) |
resets |
↻ <time> reset countdowns on the 5-hour and weekly segments |
share |
(off by default) a standalone version of the same contribution figure, labelled mine (not 5h, so it never collides with the session segment) — mine 19pp/89% by default; see §6 |
share-life |
(off by default) this session's share of all live sessions' lifetime usage, labelled life — life 22% |
sessions |
(off by default) every live session ranked by attributed 5h-window share, current one marked ▸ — A 52% · B 31% · ▸C 12% · D 5% (or ? for a session with no attribution yet) |
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.
- 8d ago First seen · 662 lines · 158 tokens per session scan A 91bc095505c1
squirrel is a skill published in the GitHub repository kubalajiri/claude-squirrel (2 stars, last pushed 14d ago), licensed MIT. It adds 158 tokens to every session and 14,351 once invoked, about $0.0008 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 skills, from other repositories
i-have-adhd
Shape output for a reader with ADHD: lead with the next action, number multi-step work, restate state across turns, suppress tangents, give specific time estimates, make wins visible. Invoke with /i-have-adhd; stays on until "stop adhd mode".
web-design-guidelines
Review UI code for Web Interface Guidelines compliance. Use when asked to "review my UI", "check accessibility", "audit design", or "vs best practices".
my-adhd-caveman
Self-contained output mode: ADHD actable structure (action first, numbered steps, one next action, one-line progress) plus caveman diction (drop fluff/articles, fragments, fewer tokens). Use when the user says "adhd-caveman", "my-adhd-caveman", or invokes /my-adhd-caveman. Needs no other skills.
idle-intel
Turn Claude Code's wait-state status line into useful micro-briefings about the current project (security vulns, outdated dependencies, unpushed commits, stale branches, TODO/FIXME debt) instead of a generic spinner. Use this skill whenever the user wants to set up idle-intel, customize their Claude Code status line…
hush
This skill should be used when the user asks to "hush", "be quiet", "stop speaking", "silence narrator", "narrator hush", "shut up", or wants to immediately stop all current and queued speech output.
off
This skill should be used when the user asks to "disable narrator", "turn off voice", "stop narrator", "disable voice output", "narrator off", "mute narrator", or wants to deactivate text-to-speech output.