retrospective
Pass
Audited by Gen Agent Trust Hub on Nov 6, 2026
Risk Level: SAFEDYNAMIC_CONTEXT_INJECTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [DYNAMIC_CONTEXT_INJECTION]: The skill uses the
!syntax in the frontmatter context field to executegit logat load time. This is used to provide historical context for the retrospective and does not access sensitive credentials, configuration files, or perform network operations.\n- [INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted data from external sources including git logs and code comments (via Grep for TODO/FIXME patterns) to generate its reports.\n - Ingestion points: Data enters the agent context via
git login the frontmatter, reading files from theproduction/directory, and Grep scans across the codebase in SKILL.md.\n - Boundary markers: No explicit boundary markers or instructions to ignore embedded commands in the ingested data were found.\n
- Capability inventory: The skill is authorized to use the
Writetool to save the generated retrospective to the filesystem.\n - Sanitization: There is no evidence of data sanitization or escaping of external content before it is interpolated into the report generation prompt.
Audit Metadata