write-milestone-brief
Pass
Audited by Gen Agent Trust Hub on Sep 14, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted data from the conversation history to generate a milestone brief.
- Ingestion points: Current conversation history and the
.gsd/STATE.mdfile. - Boundary markers: Absent; the instructions do not specify explicit delimiters or warnings to ignore instructions embedded in the synthesized text.
- Capability inventory: The skill uses the
writetool to create or overwrite files in the.gsd/directory and usesmcp__github__issue_writeto interact with external GitHub APIs. - Sanitization: Absent; the skill synthesizes established information directly into markdown templates without programmatic sanitization.
- [SAFE]: The skill follows least-privilege principles by requiring explicit user confirmation before performing outward-facing actions like creating GitHub issues.
- [SAFE]: File system operations are restricted to project-specific metadata directories (
.gsd/), which is consistent with the skill's stated purpose of managing project milestones.
Audit Metadata