effect-review
Pass
Audited by Gen Agent Trust Hub on Apr 2, 2026
Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill executes
git diffcommands to discover changed files within the local repository. These commands are used correctly for the skill's primary purpose and do not incorporate unvalidated user input.\n- [PROMPT_INJECTION]: The skill processes untrusted source code which creates a surface for indirect prompt injection. This is an inherent risk of code review agents that can be mitigated by agent-level guardrails.\n - Ingestion points: Code content from files identified via
git diffinSKILL.md.\n - Boundary markers: Absent; code content is passed to sub-agents without specific delimiters or instructions to ignore embedded commands.\n
- Capability inventory: The main skill can execute local shell commands (
git), and sub-agents have read access to the repository via the Agent tool.\n - Sanitization: No sanitization or escaping of ingested code content is performed before analysis.
Audit Metadata