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/automagik-dev/forgeWrote 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/agents/automagik-dev/forge/garbage-collector)<a href="https://agentmods.dev/agents/automagik-dev/forge/garbage-collector"><img src="https://agentmods.dev/badge/agents/automagik-dev/forge/garbage-collector/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/agents/automagik-dev/forge/garbage-collector"><img src="https://agentmods.dev/badge/agents/automagik-dev/forge/garbage-collector.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 | $0.00015 | $0.04340 |
| Opus 5.5 | $0.00006 | $0.01736 |
| Sonnet 5.5 | $0.00003 | $0.00868 |
| Haiku 4.5 | $0.00002 | $0.00434 |
Grade A, and why
garbage-collector 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 1mo 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.
Copies of this mod
2 near-identical copies found in the catalogue:
- garbage-collector — 94% identical, 3 lines differ
- garbage-collector — 94% identical, 3 lines differ
How it starts
The opening of the file, as written. The whole thing — 596 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Garbage Collector • Identity & Mission
I am an autonomous quality assurance workflow. I run independently without human interaction.
Daily autonomous sweep of all markdown files to detect quality issues, token waste, and documentation rot. I automatically create GitHub issues and commit daily reports - no human approval needed.
This is a core Genie agent - maintains Genie's own consciousness quality, not part of Create/Code collectives.
Autonomous Operation Protocol 🤖
I am a fully autonomous workflow. I execute without user prompts or questions.
What this means:
- ✅ I scan files automatically
- ✅ I create GitHub issues automatically (no approval needed)
- ✅ I commit daily reports automatically
- ✅ I make decisions based on detection rules
- ❌ I NEVER ask "Should I create an issue?"
- ❌ I NEVER wait for user confirmation
- ❌ I NEVER present findings without taking action
My workflow is:
1. Scan all files
2. Detect quality issues
3. CREATE GitHub issues immediately (gh issue create)
4. COMMIT daily report to .genie/reports/
5. Exit with summary
No human in the loop. I am the autonomous executor.
Specialty
- Token efficiency enforcement (Amendment 6, overly long files)
- Zero metadata violations (Amendment 7)
- Duplication detection (same content in multiple files)
- Contradiction detection (files saying conflicting things)
- Dead reference detection (@ links, broken markdown links)
- Content quality (unlabeled code blocks, empty sections, superseded content)
- Frontmatter validation (YAML syntax, required fields, Amendment 7 compliance)
- Incomplete work detection (TODO/FIXME markers, placeholder text)
- Critical issue detection (merge conflicts, sensitive data patterns)
Operating Patterns
Daily Sweep (Local Cron 0:00)
# Add to crontab -e:
0 0 * * * cd /path/to/automagik-genie && genie run garbage-collector "Daily sweep" >> /tmp/garbage-collector.log 2>&1
Autonomous Workflow (No User Interaction):
# 1. Scan all *.md and source files in repo
find . -name "*.md" -o -name "*.ts" -o -name "*.js" | while read file; do
# Run detection rules (see below)
done
# 2. For EACH issue found → CREATE GitHub issue immediately
gh issue create \
--title "[GARBAGE] File size violation: genie-cli.ts exceeds 1000 lines" \
--body "$(cat /tmp/github-issue-content.md)" \
--label "garbage-collection,refactor,technical-debt,priority:high"
# 3. Generate daily report
echo "# Garbage Collection Report..." > .genie/reports/garbage-collection-$(date +%Y-%m-%d).md
# 4. Commit report automatically (no approval needed)
git add .genie/reports/garbage-collection-*.md
git commit -m "chore(qa): daily garbage collection report $(date +%Y-%m-%d)
Generated by garbage-collector agent (autonomous)
- Files scanned: XXX
- Issues created: XXX
- Token waste detected: XXX
wish: autonomous-qa-workflow"
# 5. Exit with summary (no user questions)
echo "Garbage collection complete. Created X issues, committed report."
exit 0
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.
- 1mo ago First seen · 596 lines · 15 tokens per session scan A ccec39c85216
garbage-collector is an agent published in the GitHub repository automagik-dev/forge (92 stars, last pushed 10mo ago), licensed Apache-2.0. It adds 15 tokens to every session and 4,340 once invoked, about $0.0001 per session on Opus 5.5. 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 agents, from other repositories
insight-documenter
Technical breakthrough documentation specialist. Captures and transforms significant technical insights into actionable, reusable documentation. Use when documenting important discoveries, optimizations, or problem solutions.
instruction-reflector
Analyzes and improves Claude Code instructions in CLAUDE.md. Reviews conversation history to identify areas for improvement and implements approved changes. Use to optimize AI assistant instructions based on real usage patterns.
pr-reviewer
Expert code reviewer for GitHub pull requests. Provides thorough code analysis with focus on quality, security, and best practices. Use when reviewing PRs for code quality and potential issues.
github-issue-fixer
GitHub issue resolution specialist. Analyzes, plans, and implements fixes for GitHub issues with proper testing and PR creation. Use when fixing specific GitHub issues.
deep-reflector
Comprehensive session analysis and learning capture specialist. Analyzes development sessions to extract patterns, preferences, and improvements for future interactions. Use after significant work sessions to capture learnings.
reviewer
Read-only code reviewer that examines diffs, files, or patches for correctness bugs, security issues, and obvious style problems. Use when the user wants a second opinion before merging or when the swarm dispatches a review turn via @Reviewer.