update

A command for updating the noisy-coding voice backend running in Docker to its latest release.

In plain words
What is it for?
Use it to pull a newer backend image, review what changed, and restart the noisy-coding container after confirming the update.
Why use it?
It checks the current version, compares release information, and preserves the backend's configuration, API key, and conversation history when the container is recreated.

Command

Install

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.

agentmods
npx agentmods add commands/noisy/noisy-coding/update
Clone the repo
git clone --depth 1 https://github.com/noisy/noisy-coding
Per session 14 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 633 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 1 finding. Scan, not verified.
Origin original No closer match found in the catalogue.
Token cost

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.

ModelPer sessionOnce invoked
Fable 5 $0.00014 $0.00633
Opus 5 $0.00007 $0.00316
Sonnet 5 $0.00003 $0.00127
Haiku 4.5 $0.00001 $0.00063

Measured yesterday against content hash efd90aad807e, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

update scanned grade A 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 yesterday.

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.

Makes network callslowCapability

Not a fault in itself. Listed so you know the mod talks to something, and to what.

3. Fetch the release notes and tell the user WHY this update is worth it BEFORE recreating anything: `gh api repos/noisy/noisy-coding/releases --paginate -q '.[0:3]'` (or `curl -s https://api.github.com/repos/noisy/noisy
commands/update.md · 25 lines

What it actually says

Update the user's noisy-coding backend. Follow these steps in order and narrate briefly:

  1. Check what is running: docker inspect noisy-coding --format '{{.Config.Image}} {{.Image}}'. If the container does not exist, stop and point the user at /noisy-coding:setup instead.
  2. Record the current image digest, then pull: docker pull noisy/noisy-coding:latest. If the digest did not change, tell the user they are already on the latest version and stop.
  3. Fetch the release notes and tell the user WHY this update is worth it BEFORE recreating anything: gh api repos/noisy/noisy-coding/releases --paginate -q '.[0:3]' (or curl -s https://api.github.com/repos/noisy/noisy-coding/releases). Quote the Highlights sections of the releases newer than what they run — those bullets are written for exactly this purpose. Ask the user to confirm the restart (it cuts any voice conversation for a few seconds).
  4. Recreate the container, preserving the config volume (API key, settings, history live there and MUST survive):
    docker rm -f noisy-coding
    docker run -d --name noisy-coding \
      -p 127.0.0.1:8765-8767:8765-8767 \
      -v noisy-coding-config:/root/.config/noisy-coding \
      --restart unless-stopped \
      noisy/noisy-coding:latest
    
  5. Verify NON-INVASIVELY, exactly like setup does: curl -s http://127.0.0.1:8765/status must return JSON with "api_key_set": true (the volume survived). Do NOT probe port 8766 with raw TCP.
  6. THE CONTAINER IS ONLY HALF THE SYSTEM — always close with the plugin check, never skip it:
    • Read the installed plugin version (claude plugin list) and compare with the version just released. If they differ, UPDATE IT NOW yourself — claude plugin update noisy-coding@noisy works from inside the session — and tell the user the one thing you cannot do for them: run /reload-plugins (or restart) in every live session, in every Claude profile they use (a second profile needs its own CLAUDE_CONFIG_DIR=<profile> claude plugin update ...).
    • If versions match, say so explicitly ("plugin already current — container was the only half to refresh"), so the user never wonders whether a step is missing.
    • Installer users (no plugin): re-run docker exec -i noisy-coding python3 /app/hooks/install.py --docker instead when the release notes mention hook changes.
  7. Remind the user to reload the dashboard tab (hard refresh — the UI and favicon are served by the new container) and re-click ENABLE TAB AUDIO if the banner reappears.
Changes

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.

  1. yesterday First seen · 25 lines · 14 tokens per session scan A efd90aad807e

Subscribe to this mod's changes

update is a command published in the GitHub repository noisy/noisy-coding (6 stars, last pushed 3d ago), licensed MIT. It adds 14 tokens to every session and 633 once invoked, about $0.0001 per session on Opus 5. A static security scan graded it A with 1 finding (makes network calls). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.