auto-review-loop
Autonomous multi-round research review loop. Repeatedly reviews using Claude Code via claude-review MCP, implements fixes, and re-reviews until positive assessment or max rounds reached. Use when user says "auto review loop", "review until it passes", or wants autonomous iterative improvement.
67
82%
Does it follow best practices?
Run evals on this skill
Adds up to 20 points to the overall score
View guide
Critical
Do not install without reviewing
Security
3 findings: 1 critical severity, 1 high severity, 1 medium severity. Installing this skill is not recommended: please review these findings carefully if you do intend to do so.
E006: Malicious code pattern detected in skill scripts.
What this means
Detected high-risk code patterns in the skill content — including its prompts, tool definitions, and resources — such as data exfiltration, backdoors, remote code execution, credential theft, system compromise, supply chain attacks, and obfuscation techniques.
Why it was flagged
This skill autonomously sends full repository artifacts (code, diffs, raw results) to an external "claude-review" backend and includes explicit instructions to perform silent file writes and remote deployments without user prompts, creating a high risk of unintended data exfiltration and unauthorized modifications.
W007: Insecure credential handling detected in skill instructions.
What this means
The skill handles credentials insecurely by requiring the agent to include secret values verbatim in its generated output. This exposes credentials in the agent’s context and conversation history, creating a risk of data exfiltration.
Why it was flagged
The prompt mandates saving and publishing external reviewer responses and verbatim repository/raw-file evidence (diffs, raw results, memory updates) into logs, which forces the agent to handle and potentially output any secrets that appear in those sources.
W013: Attempt to modify system services in skill instructions.
What this means
The skill prompts the agent to compromise the security or integrity of the user’s machine by modifying system-level services or configurations, such as obtaining elevated privileges, altering startup scripts, or changing system-wide settings.
Why it was flagged
The skill directs autonomous execution of remote commands (SSH/screen/tmux), may start network services or install MCP servers on the user's machine, and explicitly tells the agent to run Bash file-writes "silently" without asking the user — risking unwanted state changes — although it does not request sudo, create users, or modify system-level configs directly.
Low
Low-risk findings.
1 low severity finding. Worth noting, but not necessarily harmful.
W011: Third-party content exposure detected (indirect prompt injection risk).
What this means
The skill exposes the agent to untrusted, user-generated content from public third-party sources, creating a risk of indirect prompt injection. This includes browsing arbitrary URLs, reading social media posts or forum comments, and analyzing content from unknown websites.
Why it was flagged
SKILL.md’s required runtime workflow ingests reviewer-assessment free text from an “external reviewer” route via the `claude-review` MCP calls (saved verbatim in Phase B, then used for subsequent logic), and the reviewer could be driven by outsider-authored content.
- Commit
f4f20f9
- Audited
- Security analysis
Is this your skill?
If you maintain this skill, you can claim it as your own. Once claimed, you can manage eval scenarios, bundle related skills, attach documentation or rules, and ensure cross-agent compatibility.