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/adityaraj0421/gearbox/doctorgit clone --depth 1 https://github.com/Adityaraj0421/gearboxWrote 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/adityaraj0421/gearbox/doctor)<a href="https://agentmods.dev/commands/adityaraj0421/gearbox/doctor"><img src="https://agentmods.dev/badge/commands/adityaraj0421/gearbox/doctor.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 | $0.00017 | $0.03536 |
| Opus 5 | $0.00009 | $0.01768 |
| Sonnet 5 | $0.00003 | $0.00707 |
| Haiku 4.5 | $0.00002 | $0.00354 |
Grade B, and why
doctor scanned grade B with 2 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 4d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Reads agent configuration directoriesmediumAgent snooping
.claude/, .codex/, .gemini/ hold keys, settings and other credentials a mod has no legitimate need for.
grep -l "@.claude/routing.md" CLAUDE.md 2>/dev/null && echo "CLAUDE_MD_FOUND" || echo "claude_md_clean" Makes network callslowCapability
Not a fault in itself. Listed so you know the mod talks to something, and to what.
curl -s --max-time 5 "https://raw.githubusercontent.com/Adityaraj0421/gearbox/main/.claude-plugin/plugin.json" \ How it starts
The opening of the file, as written. The whole thing — 356 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Run the following checks IN ORDER. Execute each step via Bash (or by introspection where noted). Do NOT print results as you go — collect each result silently, then print only the final table at the end.
For each check, record one of: PASS, WARN, or FAIL, plus a brief evidence/fix string.
CHECK 0 — PLUGIN ROOT
echo "${CLAUDE_PLUGIN_ROOT}"
- Non-empty output AND the path exists on disk → PASS (record the path)
- Empty or path does not exist → FAIL: "plugin variable not resolving — reinstall the plugin"
CHECK 1 — DEPENDENCIES
command -v python3 && echo "python3 ok" || echo "python3 missing"
command -v git && echo "git ok" || echo "git missing"
- Both present → PASS
- python3 missing → FAIL: "Gearbox hooks are python3 scripts; install python3 — both hooks are currently failing silently"
- git missing only → WARN: "verifier baseline checks need git"
CHECK 2 — POLICY INJECTION
Introspect without reading any files: is the Gearbox tier table (T0/T1/T2 with gearbox:scout / gearbox:grunt / gearbox:builder / gearbox:architect) present in your context right now?
- Yes → PASS
- No → FAIL: "SessionStart hook not firing — check /plugin detail view for hook load errors, then restart the session"
CHECK 3 — AGENT REGISTRY
Introspect your available subagent types. Check for all five:
gearbox:scout, gearbox:grunt, gearbox:builder, gearbox:architect,
gearbox:verifier.
- 5/5 present → PASS
- Fewer than 5 → FAIL listing the missing names: "restart the session; agents only load at session start"
CHECK 4 — INSTALL SCOPE
python3 - <<'EOF'
import json, pathlib, sys
p = pathlib.Path.home() / ".claude/plugins/installed_plugins.json"
if not p.exists():
print(f"MISSING:{p}")
sys.exit(0)
data = json.loads(p.read_text())
entries = []
for key, installs in data.get("plugins", {}).items():
if "gearbox" in key.lower():
entries.extend(installs)
if not entries:
print("NOT_FOUND")
sys.exit(0)
scopes = [e.get("scope","?") for e in entries]
print("SCOPES:" + ",".join(scopes))
EOF
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.
- 4d ago First seen · 356 lines · 17 tokens per session scan B 14054dd656f9
doctor is a command published in the GitHub repository Adityaraj0421/gearbox (9 stars, last pushed 17d ago), licensed MIT. It adds 17 tokens to every session and 3,536 once invoked, about $0.0001 per session on Opus 5. A static security scan graded it B with 2 findings (reads agent configuration directories, makes network calls). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.
Other commands, from other repositories
learn
Force claude-smart to extract learnings from this session now.
init
Install the formatters this repository needs, with every command visible before it runs.
brand-generate
Generate an on-brand document from a saved Brand Profile.
genshijin-compress
Markdown・テキストファイルを原始人形式へ安全圧縮.
save
Save this conversation as a new or existing reusable context.
wp-polylang
Translate an existing WordPress site into a second language using Polylang.