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 shreyas-s-rao/claude-code-narrator --skill offgit clone --depth 1 https://github.com/shreyas-s-rao/claude-code-narratorWrote 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/shreyas-s-rao/claude-code-narrator/off)<a href="https://agentmods.dev/skills/shreyas-s-rao/claude-code-narrator/off"><img src="https://agentmods.dev/badge/skills/shreyas-s-rao/claude-code-narrator/off.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 | $0.00050 | $0.00553 |
| Opus 5 | $0.00025 | $0.00277 |
| Sonnet 5 | $0.00010 | $0.00111 |
| Haiku 4.5 | $0.00005 | $0.00055 |
Grade B, and why
off 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 8d 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.
Only for global mode: Read `~/.claude/settings.json`. If it has a `UserPromptSubmit` hook that runs `hush-on-input.sh`, remove that entire `UserPromptSubmit` entry from the `hooks` object. Preserve all other hooks. How it starts
The opening of the file, as written. The whole thing — 46 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Disable Narrator Voice Output
To disable narrator voice output:
Step 1: Determine Scope
Check if the user said "locally", "for this directory", "for this project", or similar. If so, changes apply to the local config file in the current working directory.
- Global (default):
~/.claude-code-narrator/config - Local:
<cwd>/.claude-code-narrator/config
Step 2: Update State File
Global mode (default):
Read ~/.claude-code-narrator/config. If the file exists, use the Edit tool to change the enabled= line to enabled=false. If the file does not exist, create it with the Write tool containing just enabled=false.
Local mode:
Read <cwd>/.claude-code-narrator/config. If the file exists and has an enabled= line, use the Edit tool to change it to enabled=false. If the file exists but has no enabled= line, use the Edit tool to add enabled=false. If the file does not exist, create the <cwd>/.claude-code-narrator/ directory if needed, then create the config file with the Write tool containing just enabled=false.
To remove the local override entirely (reverting to global state), delete the <cwd>/.claude-code-narrator/ directory instead.
IMPORTANT: Do NOT use sed to edit state files — use the Read and Edit tools instead.
Step 3: Remove Auto-Hush Hook (global mode only)
Only for global mode: Read ~/.claude/settings.json. If it has a UserPromptSubmit hook that runs hush-on-input.sh, remove that entire UserPromptSubmit entry from the hooks object. Preserve all other hooks.
For local mode, do NOT remove the auto-hush hook (other sessions may still need it).
IMPORTANT: Do NOT remove other hooks — only remove the narrator UserPromptSubmit hook. Read the file first.
Step 4: Confirm
For global mode, inform the user: "Narrator is now disabled. Voice output is off."
For local mode, inform the user: "Narrator is now disabled for this directory. Global setting is unchanged."
Do NOT attempt to speak this confirmation (narrator is now off).
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.
- 8d ago First seen · 46 lines · 50 tokens per session scan B be6f507fc17a
off is a skill published in the GitHub repository shreyas-s-rao/claude-code-narrator (26 stars, last pushed 3mo ago), licensed MIT. It adds 50 tokens to every session and 553 once invoked, about $0.0003 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 2026-08-30.
Other skills, from other repositories
sitrep-panel
Keep a live local HTML report updated as you work on a task — a status board, a running "what's happening now" narrative, screenshots, and a newest-first log — served on localhost so the user can watch progress in a browser tab instead of reading the transcript. Use when the user invokes /sitrep-panel, asks for a…
web-design-guidelines
Review UI code for Web Interface Guidelines compliance. Use when asked to "review my UI", "check accessibility", "audit design", or "vs best practices".
smart-defaults
Hace que una web arranque en el tema (claro/oscuro) y el idioma correctos según las preferencias del sistema/navegador del visitante, sin parpadeo y sin forzar redirecciones. Úsalo al crear el arranque (boot) de cualquier landing o app web, al añadir un selector de tema o de idioma, o cuando el usuario pida "que la…
motion-dev-animations
Creates 120fps GPU-accelerated animations with Motion.dev (Framer Motion successor) for React, Next.js, Svelte, and Astro projects. Use when user requests animation, motion, scroll effects, parallax, hero animations, gestures, drag interactions, spring physics, whileHover effects, whileInView animations, animated UI…
first-principles-thinking
Use when approaching ANY task - bugs, features, refactoring, debugging, code review, architecture decisions. Enforces decomposition to fundamental truths before acting, prevents pattern-matching shortcuts and assumption-driven solutions.
scope-lock
ALWAYS invoke before writing any code after a plan or spec is approved. Do NOT skip even if the plan is already in context. Triggers on: plan approval, spec approval, task kickoff, implementation start, '/scope-lock', 'set up the contract', 'lock scope'. Generates SCOPE.md boundary contract from the plan, enforces…