blast-radius
Pass
Audited by Gen Agent Trust Hub on Oct 2, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted external data including code diffs, pull request descriptions, and commit messages, creating a potential vector for indirect prompt injection.
- Ingestion points: The skill instructions in SKILL.md direct the agent to read diffs, symbols, PR descriptions, and commit messages to understand changes and intent.
- Boundary markers: No explicit prompt boundary markers or "ignore instructions" delimiters are defined for these untrusted inputs.
- Capability inventory: The skill explicitly encourages running scripts or tests (SKILL.md) to prove safety facts and supports spawning subagents to parallelize analysis.
- Sanitization: The skill includes a security-positive instruction to "strip anything private before it goes anywhere public," which helps mitigate accidental data exposure.
Audit Metadata