# Agent Diagnose

> >

- **Type:** Skill
- **Install:** `agentstack add skill-timeyour-agentskills-audit-collection-agent-diagnose`
- **Verified:** Yes — security-reviewed for prompt injection and unsafe behavior
- **Seller:** [timeyour](https://agentstack.voostack.com/s/timeyour)
- **Installs:** 0
- **Category:** [Web & Browser](https://agentstack.voostack.com/c/web-and-browser)
- **Latest version:** 0.1.0
- **License:** MIT
- **Upstream author:** [timeyour](https://github.com/timeyour)
- **Source:** https://github.com/timeyour/agentskills-audit-collection/tree/main/.claude/skills/agent-diagnose
- **Website:** https://github.com/timeyour/agentskills-audit-collection

## Install

```sh
agentstack add skill-timeyour-agentskills-audit-collection-agent-diagnose
```

Requires the [AgentStack CLI](https://agentstack.voostack.com/docs/cli). Works with Claude Code, Cursor, and any MCP-compatible agent.

## About

# Agent Diagnose

Use this skill when the user wants to test whether an AI agent, prompt workflow, website-audit agent, or coding agent behaves reliably under pressure.

## Do

1. Identify the agent's claimed role, permissions, tools, and boundaries.
2. Build adversarial probes for instruction hierarchy, prompt injection, role confusion, hidden drift, authority escalation, and unverifiable claims.
3. Test auditability: can the agent show evidence, locations, logs, and source/live separation?
4. Test consistency: does it apply the same rules across similar cases?
5. Test self-correction: does it update guardrails after repeated failures?
6. Produce severity-ranked findings and fixes.
7. Use `S0-S4` severity for agent reliability impact.

## Reference

Read `references/adversarial-checklist.md`.

## Output

- Probe matrix.
- Behavioral findings.
- Evidence gaps.
- Drift findings.
- S0-S4 severity.
- Reproduction probes.
- Regression checks.
- Guardrail recommendations.
- Final reliability verdict.

## Source & license

This open-source skill is cataloged on AgentStack and links to its original source — we do not rehost the code.

- **Author:** [timeyour](https://github.com/timeyour)
- **Source:** [timeyour/agentskills-audit-collection](https://github.com/timeyour/agentskills-audit-collection)
- **License:** MIT
- **Homepage:** https://github.com/timeyour/agentskills-audit-collection

Install and usage instructions live in the source repository linked above.

## Pricing

- **Free** — Free

## Security capabilities

Automated source analysis of v0.1.0 — what this tool can access:

- **Network access:** no
- **Filesystem access:** no
- **Shell / process execution:** no
- **Environment & secrets:** no
- **Dynamic code execution:** no

*"Yes" means the capability is present in the source — more access means more to trust, not that it is unsafe.*


## Versions

- **0.1.0** — security scan: passed — Imported from the upstream source.

## Links

- Listing page: https://agentstack.voostack.com/l/skill-timeyour-agentskills-audit-collection-agent-diagnose
- Seller: https://agentstack.voostack.com/s/timeyour
- Browse the marketplace: https://agentstack.voostack.com/browse

---
Listed on AgentStack — the marketplace for AI agent skills and MCP servers. Every listing is security-reviewed. Creators keep 70%.