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 skills/blake-simpson/belmont/cleanupnpx skills add blake-simpson/belmont --skill cleanupgit clone --depth 1 https://github.com/blake-simpson/belmontWrote 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/blake-simpson/belmont/cleanup)<a href="https://agentmods.dev/skills/blake-simpson/belmont/cleanup"><img src="https://agentmods.dev/badge/skills/blake-simpson/belmont/cleanup.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 | $0.00035 | $0.03728 |
| Opus 5 | $0.00017 | $0.01864 |
| Sonnet 5 | $0.00007 | $0.00746 |
| Haiku 4.5 | $0.00003 | $0.00373 |
Grade A, and why
cleanup 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 5d 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 — 408 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Belmont: Cleanup
You are cleaning up accumulated Belmont state to reduce input token bloat. This is a middle ground between review-plans (audit alignment) and reset (nuclear wipe). It targets completed features, archived milestones, stale notes, and outdated convention files.
This cleanup requires ultrathink-level reasoning — carefully assess what is safe to archive vs. what the user still needs for active context.
Purpose
Over time, Belmont projects accumulate completed feature directories (PRD, PROGRESS, TECH_PLAN, MILESTONE-*.done.md), growing NOTES.md files, and stale conventions in CLAUDE.md/AGENTS.md. Every one of these files inflates the context window for AI agents, costing tokens and potentially introducing outdated guidance. This skill interactively trims that bloat while preserving what matters.
Critical Rules
- Interactive — every item requires explicit user choice. Never bulk-act without consent. Present each feature, each file, each finding individually.
- Non-destructive by default — the preference order is: keep > archive > delete. Default suggestion is always "keep" unless the user chooses otherwise.
- Scoped to state files — modifies
.belmont/, audits CLAUDE.md and AGENTS.md, suggests tool-dir cleanup. Does NOT modify source code. - Skip gracefully — if a category has nothing to clean, skip it silently.
Forbidden Actions
- DO NOT modify source code files
- DO NOT run builds, tests, or compilation commands
- DO NOT delete files without explicit user confirmation per item
- DO NOT auto-modify tool directories (
.claude/,.codex/,.cursor/, etc.) — only suggest - DO NOT pressure the user toward archiving — completed features may still be useful context
Allowed Actions
- DO read all
.belmont/state files, CLAUDE.md, AGENTS.md, tool directory structures - DO present cleanup options interactively via
AskUserQuestion - DO archive completed feature directories (compress verbose files into slim summaries)
- DO remove MILESTONE-*.done.md files based on user choice
- DO edit NOTES.md, CLAUDE.md, AGENTS.md based on user decisions
- DO commit
.belmont/changes after cleanup
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.
- 5d ago First seen · 408 lines · 35 tokens per session scan A e017ae1fb926
cleanup is a skill published in the GitHub repository blake-simpson/belmont (4 stars, last pushed 17d ago), licensed Apache-2.0. It adds 35 tokens to every session and 3,728 once invoked, about $0.0002 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
flow-nexus-swarm
Cloud-based AI swarm deployment and event-driven workflow automation with Flow Nexus platform.
agent-crdt-synchronizer
Agent skill for crdt-synchronizer - invoke with $agent-crdt-synchronizer.
agent-topology-optimizer
Agent skill for topology-optimizer - invoke with $agent-topology-optimizer.
agent-release-swarm
Agent skill for release-swarm - invoke with $agent-release-swarm.
agent-v3-performance-engineer
Agent skill for v3-performance-engineer - invoke with $agent-v3-performance-engineer.
agent-workflow-automation
Agent skill for workflow-automation - invoke with $agent-workflow-automation.