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/noisy/noisy-coding/clean-install-resetnpx skills add noisy/noisy-coding --skill clean-install-resetgit clone --depth 1 https://github.com/noisy/noisy-codingWhat 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 | $0.00089 | $0.00955 |
| Opus 5 | $0.00044 | $0.00477 |
| Sonnet 5 | $0.00018 | $0.00191 |
| Haiku 4.5 | $0.00009 | $0.00096 |
Grade B, and why
clean-install-reset scanned grade B 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 3d 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.
settings afterwards (`grep -i noisy ~/.claude/settings.json`). How it starts
The opening of the file, as written. The whole thing — 63 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Clean-slate reset for install testing
Purpose: bring the machine to the state of a NEW user, so the plugin install
path (marketplace add → plugin install → /noisy-coding:setup) can be
tested honestly. This skill is a checklist and a conversation script, not a
delete script — AUDIT first, show the user a table of findings, ASK about the
destructive choices, and back up whatever you edit (one local directory,
e.g. .cleanup-backup-<date>/, kept out of git).
Expect the permission classifier to block some steps (editing settings files
from scripts, docker rm): fall back to the Edit tool for files, and hand the
user a ready !-prefixed command for the docker parts.
Where artifacts hide (audit ALL of these)
~/.claude/settings.json— installer-era hooks (docker exec -i noisy-coding …) andenabledPlugins/extraKnownMarketplacesentries.~/.claude.json— a user-scopemcpServers.noisy-codingentry (from the manualclaude mcp addpath) and deadprojectsentries (e.g. the oldgrok-voice-mcpcheckout).~/.claude/plugins/—installed_plugins.json,known_marketplaces.json(marketplacenoisy),cache/noisy/. Preferclaude plugin uninstall/claude plugin marketplace removeover hand-editing when available.- Plugin hooks live inside the plugin cache — they vanish with the plugin;
no separate cleanup, but VERIFY nothing noisy-related remains in the main
settings afterwards (
grep -i noisy ~/.claude/settings.json). ~/.claude/skills/— hand-copiedgrok-voice*relics.~/.config/noisy-coding/— stalerewake-*.lockfiles andsessions.jsonare always safe to delete;credentials.json,history.json,character.json,settings.jsonare USER DATA — ask (see below).~/.config/grok-voice/— pre-rename leftovers, safe to delete.- Docker: container
noisy-coding, volumenoisy-coding-config, legacy images (grok-voice*, oldnoisy/noisy-coding).docker rmneeds the user; give the exact command. - Secondary Claude profiles the user may have (e.g.
~/.claude-personal) — same checks as 1–3. Ask which profiles are in scope.
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.
- 3d ago First seen · 63 lines · 0 tokens per session scan B 93bb287af564
clean-install-reset is a skill published in the GitHub repository noisy/noisy-coding (6 stars, last pushed 4d ago), licensed MIT. It adds 89 tokens to every session and 955 once invoked, about $0.0004 per session on Opus 5. A static security scan graded it B with 1 finding (reads agent configuration directories). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.
Other skills, from other repositories
start-talking
Start and run an explicit, interactive Cadence Code conversation with Codex or Antigravity using fully local speech input and output. Use only when the user explicitly invokes $start-talking, /start-talking, or asks to start talking with Cadence Code.
jump-in
Interrupt an active Cadence Code conversation and add fresh spoken guidance to the current Codex or Antigravity task. Use only when the user explicitly invokes $jump-in or /jump-in after stopping the current host turn.
voice-settings
Choose Cadence Code's local speech and transcription models from the Codex or Antigravity UI. Use only when the user explicitly invokes $voice-settings, /voice-settings, or asks to open Cadence Code settings.
wrap-up
End an active Cadence Code conversation cleanly and release its local speech models. Use only when the user explicitly invokes $wrap-up, /wrap-up, or asks to wrap up the voice conversation.
speech-expert
Expert guide for speech-mcp - TTS, local/streaming STT, barge-in, wake word, voice command bus, and provider selection.
live-voice-agent
Use Live Voice Agent for gated Korean/English voice. Wake is 헤이 자비스.