gathering-growth

Pass

Audited by Gen Agent Trust Hub on Oct 9, 2026

Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill utilizes local shell commands including git log to extract recent development context and the GitHub CLI (gh) to create issues for content tracking. These operations are within the expected scope of a content planning tool and rely on the user's existing local authentication.
  • [INDIRECT_PROMPT_INJECTION]: The workflow incorporates an attack surface for indirect prompt injection by ingesting data from external URLs (during SEO audits) and project history. This risk is mitigated by design through mandatory human-in-the-loop checkpoints that require manual approval of all subagent outputs before the process proceeds to subsequent stages or writes data.
  • [SAFE]: The orchestration logic follows security best practices by isolating subagents (Hummingbird, Squirrel, Firefly, Wren), ensuring each model only receives the minimal context required for its specific task rather than full project access.
Audit Metadata
Risk Level
SAFE
Analyzed
Oct 9, 2026, 01:31 AM