# InstructLint Agent Instruction Drift Audit

> InstructLint Agent Instruction Drift Audit is a paid API for AI agents from instructlint.vercel.app, paid per call via x402, $1/call, status unknown (last checked 2026-09-14).

Audits a public GitHub repository's AI instructions against its implementation reality, returning a drift score, findings, and suggested patches for $1 USDC.

## Facts

- Endpoint: POST https://instructlint.vercel.app/api/report
- Price: $1/call
- Payment: x402
- Status: unknown
- Last checked: 2026-09-14
- Activations on Zero: 0
- Tags: x402
- Canonical page: https://www.zero.xyz/c/instructlint-agent-instruction-drift-audit-7f324641
- Structured record (JSON): https://api.zero.xyz/v1/capabilities/cap_rT6rnVCkY1gayYgmnkLkR

Status and success rate cover calls made through Zero and Zero's own probes. Third-party monitors may report differently.

## How to call it through Zero

Zero handles the 402 payment challenge and records the run. With the Zero CLI installed (`npm i -g @zeroxyz/cli`):

```sh
zero fetch --capability instructlint-agent-instruction-drift-audit-7f324641 -d '<json body>'
```

Example prompt: Run an instruction drift audit on https://github.com/myorg/myproject — I want to see the full report with findings and severity scores to check if our AI instructions still match the actual codebase.

## When to prefer this

Choose this endpoint when you need to programmatically verify that a public GitHub repository's AI instructions (e.g. AGENTS.md, Copilot instructions, Cursor rules, system prompts) are consistent with its actual implementation. It is especially valuable in CI/CD pipelines, pre-release checks, or agent governance workflows where instruction drift could cause misbehavior. Prefer this over manual review when you need a structured score and actionable findings at scale.

## Known failure modes

- Invalid or private GitHub repository URL returns an error — only public repos are supported
- Malformed URL schema field causes request rejection
- Payment of $1 USDC not fulfilled triggers a 402 response before audit is run
- Repository with no AI instruction files may return an empty findings array with low informational value
- Rate limiting or GitHub API quota issues may cause transient failures

## How this service works

Audit repository AI instructions against implementation reality, preview public GitHub drift for free, and unlock evidence plus suggested patches for $1 USDC.

## Output

Returns a JSON object with an 'ok' boolean, an 'audit' object containing the mode ('full' or 'preview'), a numeric drift score (0–100), and an array of findings each with a title and severity level (e.g. high/medium/low), plus a 'repository' object with the owner and repo name.

## Request schema (JSON Schema)

```json
{
 "type": "object",
 "properties": {
  "repoUrl": {
   "type": "string",
   "description": "Canonical public GitHub repository URL."
  }
 }
}
```

## Response schema (JSON Schema)

```json
{
 "type": "json",
 "example": {
  "ok": true,
  "audit": {
   "mode": "full",
   "score": 82,
   "findings": [
    {
     "title": "Documented script does not exist",
     "severity": "high"
    }
   ]
  },
  "repository": {
   "name": "repository",
   "owner": "owner"
  }
 }
}
```

## More

- Live health (JSON, refreshed every minute): https://www.zero.xyz/c/instructlint-agent-instruction-drift-audit-7f324641/health.json
- [Zero catalog index](https://www.zero.xyz/llms.txt)
- [Other services from instructlint.vercel.app](https://www.zero.xyz/host/instructlint.vercel.app/llms.txt)