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/madappgang/claude-code/configgit clone --depth 1 https://github.com/MadAppGang/claude-codeWhat 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 | $0.00012 | $0.01449 |
| Opus 5 | $0.00006 | $0.00724 |
| Sonnet 5 | $0.00002 | $0.00290 |
| Haiku 4.5 | $0.00001 | $0.00145 |
Grade A, and why
autopilot-config 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 2d 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 — 233 lines — stays where its author put it; the contents beside it link to each section on GitHub.
<user_request> $ARGUMENTS </user_request>
# Check playwright and browser binaries
if ! bun pm ls 2>/dev/null | grep -q "playwright"; then
echo "Installing playwright..."
bun add playwright
npx playwright install chromium
fi
```
</step>
<step>
Check for existing configuration:
- Environment variables (LINEAR_API_KEY, etc.)
- Settings file (.claude/autopilot.local.md)
- Plugin settings in .claude/settings.json
</step>
<step>
If config exists, show current settings
</step>
</steps>
</phase>
<phase number="2" name="Validate Credentials">
<steps>
<step>
If LINEAR_API_KEY missing:
```
AskUserQuestion:
"Linear API key not found.
To get your API key:
1. Go to Linear Settings > API
2. Create a Personal API Key
3. Copy the key (starts with lin_api_)
Paste your API key:"
```
</step>
<step>
Test API connection:
```bash
bun run ${CLAUDE_PLUGIN_ROOT}/scripts/linear-client.ts test-connection
```
</step>
</steps>
</phase>
<phase number="3" name="Select Team/Project">
<steps>
<step>
Query available teams:
```bash
bun run ${CLAUDE_PLUGIN_ROOT}/scripts/linear-client.ts list-teams
```
</step>
<step>
AskUserQuestion with team list:
"Select your team:
1. {team_1_name} ({team_1_id})
2. {team_2_name} ({team_2_id})"
</step>
<step>
Query available projects in selected team
</step>
<step>
AskUserQuestion with project list (optional)
</step>
</steps>
</phase>
<phase number="4" name="Configure Tag Mappings">
<steps>
<step>
Show default tag mappings:
"@frontend -> /dev:feature
@backend -> /dev:implement
@debug -> /dev:debug
@test -> /dev:test-architect
@review -> /commit-commands:commit-push-pr
@refactor -> /dev:implement
@research -> /dev:deep-research
@ui -> /dev:ui"
</step>
<step>
AskUserQuestion:
"Would you like to customize tag mappings?
1. Use defaults
2. Add custom mappings
3. Edit existing mappings"
</step>
<step>
If customizing, collect custom mappings
</step>
</steps>
</phase>
<phase number="5" name="Configure Proof Settings">
<steps>
<step>
AskUserQuestion:
"Configure proof-of-work settings:
1. Auto-approval confidence threshold (default: 95%)
2. Max feedback iterations (default: 5)
3. Require screenshots for UI tasks (default: yes)
4. Require test coverage (default: 80%)
Press Enter to accept defaults or type values:"
</step>
</steps>
</phase>
<phase number="6" name="Save Configuration">
<steps>
<step>
Check if .claude/autopilot.local.md already exists:
```bash
test -f .claude/autopilot.local.md && echo "exists" || echo "not found"
```
If exists, ask for confirmation:
```
AskUserQuestion:
"Configuration file .claude/autopilot.local.md already exists.
Overwrite with new settings? (yes/no)"
```
Only proceed if user confirms.
</step>
<step>
Write configuration to .claude/autopilot.local.md:
```yaml
---
linear:
team_id: "{team_id}"
project_id: "{project_id}"
bot_user_id: "{bot_user_id}"
proof:
confidence_threshold: 95
max_feedback_rounds: 5
require_screenshots: true
require_coverage: 80
tag_mappings:
"@frontend":
command: "/dev:feature"
agent: "developer"
skills: ["react-typescript"]
"@backend":
command: "/dev:implement"
agent: "developer"
skills: ["golang", "api-design"]
---
# Autopilot Configuration
This file stores project-specific autopilot settings.
## Custom Notes
Add any project-specific notes here.
```
</step>
<step>
Verify configuration saved
</step>
</steps>
</phase>
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.
- 2d ago First seen · 233 lines · 12 tokens per session scan A 4030efce172c
autopilot-config is a command published in the GitHub repository MadAppGang/claude-code (279 stars, last pushed 5mo ago), licensed MIT. It adds 12 tokens to every session and 1,449 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
checklist
Generate a custom checklist for the current feature based on user requirements.
clarify
Identify underspecified areas in the current feature spec by asking up to 5 highly targeted clarification questions and encoding answers back into the spec.
specify
Create or update the feature specification from a natural language feature description.
analyze
Perform a non-destructive cross-artifact consistency and quality analysis across spec.md, plan.md, and tasks.md after task generation.
constitution
Create or update the project constitution from interactive or provided principle inputs.
converge
Assess the current codebase against the feature's spec, plan, and tasks, then append any remaining unbuilt work as new tasks to tasks.md so implement can complete it.