developer-relations-kickoff
Pass
Audited by Gen Agent Trust Hub on Nov 1, 2026
Risk Level: SAFE
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted data for context detection. 1. Ingestion points: README.md, CHANGELOG.md, and Git history (SKILL.md, Section 1). 2. Boundary markers: No specific delimiters are defined for parsed file content. 3. Capability inventory: The skill performs file writes (devrel-context.md) and executes shell-based interrogation commands. 4. Sanitization: No explicit sanitization of ingested file content is mentioned. This risk is inherent to the skill's purpose of automated context detection.\n- [SAFE]: The skill implements a robust data-handling policy in its Memory section (Section 6), which forbids storing sensitive information such as PII, moderation reports, and unannounced project details, mitigating data exposure risks.
Audit Metadata