Code Reviewer
v1.1.0Conduct rigorous, adversarial code reviews with zero tolerance for mediocrity. Default behavior is a single-model adversarial review that identifies security...
⭐ 0· 135·0 current·0 all-time
byCorbin Breton@corbin-breton
MIT-0
Download zip
LicenseMIT-0 · Free to use, modify, and redistribute. No attribution required.
Security Scan
OpenClaw
Benign
high confidencePurpose & Capability
Name/description align with the instructions. All declared requirements (none) are appropriate for a review-only skill. The SKILL.md asks for reviewing code, optional cross-model review, and logging; nothing requires unrelated credentials or system access.
Instruction Scope
The instructions expect the agent to read code artifacts/diffs and optionally spawn a second-model sub-agent (e.g., GPT-5.4) for cross-model review. This is within scope for a review skill, but --dual mode will cause additional model invocations and costs and may rely on availability of a different model family. The skill explicitly states it only reads code and does not modify source files.
Install Mechanism
No install spec and no code files — lowest-risk format. Nothing is downloaded or written to disk by the skill itself.
Credentials
The skill declares no environment variables, secrets, or config paths. SKILL.md does not request credentials or access to unrelated system config. Proportional to purpose.
Persistence & Privilege
always is false; agent invocation is allowed (normal). The skill does not request permanent presence or modifications to other skills or system-wide settings. It does mention logging the review trail when workflows call for evidence — that is a reasonable, scoped instruction.
Assessment
This is an instruction-only reviewer that appears to do what it says: read code and produce adversarial reviews. Before installing, note: (1) --dual mode will invoke an additional model family (higher cost, may require platform support); verify your agent platform can safely spawn the alternate model and that you’re comfortable with the extra API usage and cost. (2) The skill says it logs review trails when requested — confirm where logs are stored in your environment to avoid unintentionally persisting sensitive code. (3) The reviewer is adversarial by design and blunt; expect direct, critical feedback. (4) Because it reads code artifacts, ensure you grant it access only to the repositories/files you intend it to see. No credentials, installs, or file writes are requested by the skill itself.Like a lobster shell, security has layers — review code before you run it.
latestvk97bxqx7tdsd25nm3sgdan66fx83x1qm
License
MIT-0
Free to use, modify, and redistribute. No attribution required.