start-implementation

Pass

Audited by Gen Agent Trust Hub on Mar 25, 2026

Risk Level: SAFEPROMPT_INJECTION
Full Analysis
  • [PROMPT_INJECTION]: The skill exhibits a vulnerability surface for indirect prompt injection via untrusted GitHub issue content.
  • Ingestion points: The skill ingests data including issue titles, numbers, and types from GitHub to determine implementation steps (SKILL.md).
  • Boundary markers: The instructions lack explicit delimiters or 'ignore embedded instructions' warnings when processing issue data.
  • Capability inventory: The skill possesses significant capabilities including code implementation, writing issue summaries, committing changes to the repository, and closing issues.
  • Sanitization: There is no evidence of sanitization or validation of the ingested issue content before the agent performs implementation and commitment tasks.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 25, 2026, 08:49 AM