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 skills add NicoGenti/opencode-orchestrator-kit --skill dev-cleanupgit clone --depth 1 https://github.com/NicoGenti/opencode-orchestrator-kitWrote 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/nicogenti/opencode-orchestrator-kit/dev-cleanup)<a href="https://agentmods.dev/skills/nicogenti/opencode-orchestrator-kit/dev-cleanup"><img src="https://agentmods.dev/badge/skills/nicogenti/opencode-orchestrator-kit/dev-cleanup/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/skills/nicogenti/opencode-orchestrator-kit/dev-cleanup"><img src="https://agentmods.dev/badge/skills/nicogenti/opencode-orchestrator-kit/dev-cleanup.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 | $0.00074 | $0.01539 |
| Opus 5 | $0.00037 | $0.00770 |
| Sonnet 5 | $0.00015 | $0.00308 |
| Haiku 4.5 | $0.00007 | $0.00154 |
Grade C, and why
dev-cleanup scanned grade C with 1 finding 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 7d 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.
Recursive force deletehighDestructive command
rm -rf with a variable or a broad path is one typo away from removing the wrong tree.
Recycle Bin on Windows), prefer reversible deletion over `rm -rf`. If only `rm` is available, the How it starts
The opening of the file, as written. The whole thing — 132 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Dev Cleanup (Confirmation-Gated)
A confirmation-gated checklist for cleaning local dev artifacts. This skill is opt-in and applies when the user explicitly asks for cleanup, or when stale artifacts are the demonstrated source of a reproducible failure. It is NOT a general maintenance routine and it MUST NOT be invoked for ordinary tasks.
Safety Contract (Read First)
This skill is intentionally narrow because cleanup commands are some of the most common sources of unintended data loss in agentic workflows. The orchestrator MUST follow these rules without exception:
- No automatic cleanup. This skill MUST NOT execute any destructive command without an explicit, in-the-moment "yes" from the user in the root session, naming the exact paths and commands. A prior blanket "go ahead and clean up" is NOT sufficient when new paths or commands appear later.
- Default to read-only inspection first. Before recommending any cleanup, list the candidate paths, their sizes, and why each is safe or unsafe to delete. The user decides.
- One action per confirmation. Do not batch multiple destructive commands into a single yes/no. Each command gets its own confirmation listing the exact paths it touches.
- Stop on the first unexpected file. If a target directory contains anything outside the documented list (untracked source, downloaded datasets, scratch notes), stop and ask before continuing.
- Never touch secret material. Do not clean
.env,.env.*,secrets/,*.pem,*.key,credentials.json, or anything matchingsecret/credentialin its name, even if the user asks broadly to "clean everything". - Trash, do not shred. Where the platform supports it (
trashon macOS,gio trashon Linux, Recycle Bin on Windows), prefer reversible deletion overrm -rf. If onlyrmis available, the user MUST be told it is irreversible before confirmation.
When to Use
- The user explicitly asks for cleanup ("clean my build cache", "free up disk space", "reset the dev environment").
- A reproducible failure has been traced to a stale artifact (corrupted bundler cache, stale
node_modules, leftover.tsbuildinfo) and the smallest fix is to remove that artifact. - A repository's bootstrap doc (
README.md,CONTRIBUTING.md) defines a documented cleanup ritual.
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.
- 7d ago First seen · 132 lines · 74 tokens per session scan C e9807bde73fe
dev-cleanup is a skill published in the GitHub repository NicoGenti/opencode-orchestrator-kit (5 stars, last pushed 8d ago), licensed MIT. It adds 74 tokens to every session and 1,539 once invoked, about $0.0004 per session on Opus 5. A static security scan graded it C with 1 finding (recursive force delete). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-09-05.
Other skills, from other repositories
agent-self-scheduling
Schedule AI agent runs with cron, loops, or external clocks while avoiding unsafe tight autonomous timers.
cao-session-management
Interact with CAO (CLI Agent Orchestrator) — launch multi-agent sessions, check status, send follow-up instructions, unblock stuck terminals, or shut down sessions. Use when working with CAO sessions in any capacity.
workflow
A workflow skill for embedded-device projects that plans and runs building, flashing, debugging, observing, and diagnosis tasks.
ultrawork
Parallel execution engine for high-throughput task completion.
apple-notes-search
Semantic + keyword search and connection-discovery across the user's own Apple Notes via the apple-notes MCP server. Use when the user wants to find, recall, or synthesize something from their notes, or surface non-obvious bridges/related notes. macOS, on-device.
wrap-session
Wrap up the current session by updating active.md, logging the session, committing, and providing the next session's starter prompt. Use at the end of every working session.