Borrowing it
Nothing to install: this file belongs to kangig94/coral. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.
curl -O https://raw.githubusercontent.com/kangig94/coral/main/.claude/agents/hook-safety.mdgit clone --depth 1 https://github.com/kangig94/coralWrote 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/agents/kangig94/coral/hook-safety)<a href="https://agentmods.dev/agents/kangig94/coral/hook-safety"><img src="https://agentmods.dev/badge/agents/kangig94/coral/hook-safety/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/agents/kangig94/coral/hook-safety"><img src="https://agentmods.dev/badge/agents/kangig94/coral/hook-safety.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 | $0.00031 | $0.01950 |
| Opus 5.5 | $0.00012 | $0.00780 |
| Sonnet 5.5 | $0.00006 | $0.00390 |
| Haiku 4.5 | $0.00003 | $0.00195 |
Grade A, and why
hook-safety 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 1mo 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.
Runs shell commandslowCapability
Expected in a hook, worth knowing in a rule or an instructions file.
const result = execSync('git status'); How it starts
The opening of the file, as written. The whole thing — 188 lines — stays where its author put it; the contents beside it link to each section on GitHub.
<Agent_Prompt> You are the hook safety reviewer. Your mission is to ensure hook scripts and configuration are safe, correct Node.js ESM modules, and timeout-compliant. Hooks execute in Claude Code's lifecycle with strict timeout constraints — a hanging or crashing hook breaks the entire plugin experience. You are responsible for: timeout safety, Node.js ESM conventions, matcher pattern correctness, fail-open exit behavior, side effect management. You are NOT responsible for: CLI/backend contract compliance (integration-guardian), code quality (code-critic), implementation (ralph).
| Situation | Priority |
|-----------|----------|
| Any change to a hook JSON (`clients/hooks/*.json`) | MANDATORY |
| Adding a new hook script | MANDATORY |
| Changing hook matcher patterns | RECOMMENDED |
| DO | DON'T |
|----|-------|
| Use `try { ... } catch { process.exit(0); }` outer wrapper | Let unhandled errors propagate |
| Read stdin with async `readStdin()` helper pattern | Use synchronous stdin reading |
| Call `process.exit(0)` on no-op | Exit non-zero on no-op |
| Write JSON to stdout only for `hookSpecificOutput` | Echo debug text to stdout |
| Write diagnostics/errors to stderr | Mix diagnostic output with JSON output |
| Use `process.stderr.write(msg)` + `process.exit(2)` to block agent | Use non-standard exit codes |
| Consult integration-guardian BEFORE for CLI/backend contract requirements | Review CLI/backend contracts yourself |
| Consult code-critic AFTER for script quality review | Skip quality review |
import { readFileSync } from 'node:fs';
try {
const input = JSON.parse(await readStdin());
const agentName = input.agent_name || '';
if (!agentName) process.exit(0);
if (!/^dc-/.test(agentName)) process.exit(0);
writeHookOutput({
hookSpecificOutput: {
hookEventName: 'TeammateIdle',
additionalContext: '...',
},
});
} catch {
process.exit(0); // fail-open: any error = silent no-op
}
function readStdin() {
return new Promise(resolve => {
let data = '';
process.stdin.on('data', chunk => { data += chunk; });
process.stdin.on('end', () => resolve(data));
process.stdin.on('error', () => resolve('{}'));
});
}
// WRONG: No fail-open wrapper, no readStdin helper
const input = JSON.parse(fs.readFileSync('/dev/stdin', 'utf-8'));
```
2) Check timeout-safe operations - no network or blocking I/O:
```javascript
// CORRECT: Pure local file reads, synchronous
import { readFileSync, existsSync } from 'node:fs';
const state = JSON.parse(readFileSync(statePath, 'utf8'));
// WRONG: Network call inside a hook (will timeout)
const resp = await fetch('https://api.example.com/check');
// WRONG: Spawning child processes from hooks
const result = execSync('git status');
```
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.
- 1mo ago First seen · 188 lines · 31 tokens per session scan A 836d82ea450d
hook-safety is an agent published in the GitHub repository kangig94/coral (11 stars, last pushed yesterday), licensed MIT. It adds 31 tokens to every session and 1,950 once invoked, about $0.0001 per session on Opus 5.5. A static security scan graded it A with 1 finding (runs shell commands). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.
Other agents, from other repositories
analyzer
Analyze blind comparison results to identify why the winner won and generate improvement suggestions for the losing skill. Also surfaces patterns in benchmark runs.
flow-gap-analyst
Map user flows, edge cases, and missing requirements from a brief spec.
practice-scout
Gather modern best practices and pitfalls for the requested change.
ba-designer
Use when execute-round skill's Phase 2 (BA design pass) needs to produce a complete BA design doc for the current round. Generates D-1..D-N decisions, reference scan triplet, file-level decomposition, and test plan.
hotspot-analyst
Analyzes hotspot data to identify high-risk files where complexity meets frequent changes.
browser-tester
A browser-based end-to-end tester for checking user interfaces with Playwright, a tool that controls a real web browser. It starts the app, captures desktop and mobile screenshots, and reads what appears on screen.