Skills
skills/donchitos/claude-code-game-studios/milestone-review/Gen Agent Trust Hub

milestone-review

Pass

Audited by Gen Agent Trust Hub on May 13, 2026

Risk Level: SAFEPROMPT_INJECTION
Full Analysis
  • [PROMPT_INJECTION]: The skill exhibits an indirect prompt injection vulnerability surface by ingesting and processing untrusted data from the project environment.
  • Ingestion points: The skill reads milestone definitions from production/milestones/, sprint reports from production/sprints/, and scans the codebase for TODO, FIXME, and HACK markers using the Grep tool (Phase 1 and 2).
  • Boundary markers: Absent. There are no instructions or delimiters defined to ensure the agent ignores or isolates potential instructions embedded within the ingested data.
  • Capability inventory: The skill utilizes Read, Glob, Grep, Write, Task, and AskUserQuestion. The Task tool is used to invoke a 'producer' agent, which is passed data derived from the untrusted sources.
  • Sanitization: No sanitization, escaping, or validation of the external content is performed before it is interpolated into the milestone review or passed to subsequent tasks.
Audit Metadata
Risk Level
SAFE
Analyzed
May 13, 2026, 11:50 AM