claude-cli-agent — independently scanned and version-tracked by SaferSkills.
SaferSkills independently audited claude-cli-agent (Agent Skill) and scored it 87/100 (green). The audit ran 55 deterministic rules across Security, Supply Chain, Maintenance, Transparency, and Community; it found 0 high-severity and 3 lower-severity findings. The full rule-by-rule trace and per-finding evidence are below. Free, methodology-open.
Findings & checks · 3 flagged
The text {match} tells the agent to skip the normal "ask the user first" gate. Used adversarially it removes the human-in-the-loop check before destructive or sensitive actions, turning a normally-gated agent into a fire-and-forget executor.
Every scanned point with the score it earned and what moved between them.
First recorded scan — no prior version to compare against.
The primary manifest — the file an agent reads to learn what this artifact does.
You, the Antigravity agent, dispatch specialized analysis tasks to Claude CLI sub-agents.
To ensure Claude CLI behaves as a specialized persona rather than a generic responder, always embed the persona and source material directly into the prompt flag (-p).
claude --model haiku-4.5 -p "$(cat agents/persona.md)
---SOURCE CODE---
$(cat target.py)
---INSTRUCTION---
Perform a full code review. Use severity levels: 🔴 CRITICAL, 🟡 MODERATE, 🟢 MINOR.
You are operating as an isolated sub-agent.
Do NOT use tools. Do NOT access filesystem." > review.mdrun_agent.py (Cross-Platform)For reusable sub-agent execution, use the provided Python orchestrator which handles temp file assembly and prompt concatenation reliably across Windows, macOS, and Linux.
# Location: plugins/cli-agents/scripts/run_agent.py
python ./scripts/run_agent.py <PERSONA_FILE> <INPUT_FILE> <OUTPUT_FILE> "<INSTRUCTION>"python ./scripts/run_agent.py agents/security-auditor.md target.py security.md \
"Find vulnerabilities. Use severity levels: 🔴 CRITICAL, 🟡 MODERATE, 🟢 MINOR."agents/)These personas are mirrored from the Gemini and Copilot plugins to ensure consistent "Agentic" analysis across the ecosystem.
| Persona | Use For |
|---|---|
security-auditor.md | Red team, vulnerability scanning, threat modeling |
refactor-expert.md | Optimizing code for readability, performance, and DRY |
architect-review.md | Assessing system design, modularity, and complexity |
For rapid, cost-effective analytical sub-agent tasks, always specify --model haiku-4.5. It provides the best latency for "Inner Loop" code reviews.
If your prompt starts with YAML frontmatter (e.g., ---), some shell parsers might misinterpret the flag. Always prepend a newline to the prompt string when passing it to -p. (Note: The run_agent.py script handles this automatically).
Always add these instructions to your dispatch prompt to prevent the sub-agent from attempting to use external tools:
"You are operating as an isolated sub-agent. Do NOT use tools. Do NOT access filesystem. Only use the provided input."
python ./scripts/run_agent.py agents/refactor-expert.md target.py output.md "Refactor this code."~30 seconds. Free. No account. Every finding cites a rule and a line of evidence.