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/gf-labs/claude-toolboxWrote 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/gf-labs/claude-toolbox/doctor)<a href="https://agentmods.dev/commands/gf-labs/claude-toolbox/doctor"><img src="https://agentmods.dev/badge/commands/gf-labs/claude-toolbox/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 5.1 | $0.00007 | $0.03270 |
| Opus 5 | $0.00003 | $0.01635 |
| Sonnet 5 | $0.00001 | $0.00654 |
| Haiku 4.5 | $0.00001 | $0.00327 |
Grade C, and why
doctor scanned grade C with 2 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 8d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Reads agent configuration directoriesmediumAgent snooping
.claude/, .codex/, .gemini/ hold keys, settings and other credentials a mod has no legitimate need for.
cat ~/.claude/settings.json 2>/dev/null || echo "NOT FOUND" Reads MCP configurationmediumAgent snooping
mcp.json carries server URLs and auth tokens; reading it lets a mod discover and abuse other integrations.
cat .mcp.json 2>/dev/null || echo "NOT FOUND" How it starts
The opening of the file, as written. The whole thing — 347 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Collect context
Run each command below now before proceeding. Store results mentally.
Today's date, repo, CLAUDE.md:
echo "DATE:" && date +%Y-%m-%d
echo "REPO:" && (basename "$(git rev-parse --show-toplevel 2>/dev/null)" 2>/dev/null || echo "(not a git repo)")
echo "CLAUDE_MD:" && (test -f CLAUDE.md && echo "present" || echo "MISSING")
Commands and scripts:
echo "COMMANDS:" && (ls commands/*.md 2>/dev/null || echo "none")
echo "SCRIPTS:" && (ls scripts/*.py 2>/dev/null || echo "none")
hooks.json:
cat hooks/hooks.json 2>/dev/null || echo "NOT FOUND"
plugin.json:
cat .claude-plugin/plugin.json 2>/dev/null || echo "NOT FOUND"
pyproject.toml:
test -f pyproject.toml && echo "EXISTS" || echo "MISSING"
JSON validity:
python3 -c "
import json, glob
files = []
for pattern in ['*.json', '.claude/*.json', '.claude-plugin/*.json', 'hooks/*.json']:
files.extend(glob.glob(pattern))
files = sorted(set(files))
if not files:
print('no JSON files found')
else:
for f in files:
try:
json.load(open(f))
print('OK: ' + f)
except Exception as e:
print('ERROR: ' + f + ': ' + str(e))
" 2>/dev/null || echo "python3 not available"
Python syntax:
python3 -c "
import py_compile
from pathlib import Path
files = [str(f) for f in Path('.').rglob('*.py')
if not any(p in f.parts for p in ['.git', '.venv', '__pycache__', 'node_modules'])]
if not files:
print('no Python files found')
else:
for f in sorted(files):
try:
py_compile.compile(f, doraise=True)
print('OK: ' + f)
except py_compile.PyCompileError as e:
print('ERROR: ' + f + ': ' + str(e))
" 2>/dev/null || echo "python3 not available"
.gitignore:
cat .gitignore 2>/dev/null || echo "not found"
Invocation mode:
echo "${ARGUMENTS:-}"
Scope:
python3 -c "
import os, sys
sys.path.insert(0, os.environ.get('CLAUDE_TOOLBOX_ROOT', '') + '/scripts')
from _scope import get_scope
mode, data, cwd = get_scope()
if mode == 'single':
print(f'SINGLE {cwd.name} ({data})')
elif mode == 'parent':
print(f'PARENT {cwd} — {len(data)} child projects:')
for _key, path in data:
print(f' {path}')
else:
print('GLOBAL')
" 2>/dev/null || echo "GLOBAL"
Global settings:
cat ~/.claude/settings.json 2>/dev/null || echo "NOT FOUND"
Project settings:
cat .claude/settings.json 2>/dev/null || echo "NOT FOUND"
Project MCP config:
cat .mcp.json 2>/dev/null || echo "NOT FOUND"
Global commands:
ls ~/.claude/commands/ 2>/dev/null || echo "EMPTY"
Global agents:
ls ~/.claude/agents/ 2>/dev/null || echo "NONE"
Current repo:
basename "$(git rev-parse --show-toplevel 2>/dev/null)" 2>/dev/null || echo "(not a git repo)"
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 · 347 lines · 7 tokens per session scan C 1ee9f65119d3
doctor is a command published in the GitHub repository gf-labs/claude-toolbox (2 stars, last pushed 2d ago), licensed MIT. It adds 7 tokens to every session and 3,270 once invoked, about $0.0000 per session on Opus 5. A static security scan graded it C with 2 findings (reads agent configuration directories, reads mcp configuration). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.
Other commands, from other repositories
parallax-status
Summarize the current session and run package/native lifecycle diagnostics.
flutter-build
Fix Dart analyzer errors and Flutter build failures incrementally. Invokes the dart-build-resolver agent for minimal, surgical fixes.
quick-fix
Run the implement→review subagent loop without the planning phase — for straightforward fixes with a known cause and one obvious approach, regardless of how many files the change spreads across. Skips the spec gate, worktree gate, and finalize ceremony (implementation log, docs, signals) that /subagent-implementation…
debug
Systematic debugging with hypotheses and evidence gathering.
doctor
Run the specrails health check to validate that all prerequisites are correctly configured for this repository.
debug
A debugging command that sends an error, log, or screenshot to a debugger agent for analysis. The debugger reports confirmed facts separately from possible explanations and does not modify files.