monthly-report-generator

Pass

Audited by Gen Agent Trust Hub on Jul 21, 2026

Risk Level: SAFEPROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
  • [PROMPT_INJECTION]: The skill exhibits a surface for indirect prompt injection. It delegates data gathering to sub-agents that ingest untrusted content from Git commit messages and GitHub Pull Request descriptions. Ingestion points: Commit logs and PR bodies retrieved via gh pr list (Workflow Step 1). Boundary markers: Absent; there are no instructions to the sub-agents to ignore or delimit embedded instructions within the ingested data. Capability inventory: The skill has the ability to generate HTML files and perform git push operations to update project wikis (Workflow Step 5). Sanitization: None; the skill interpolates external descriptions directly into the report compilation process.
  • [COMMAND_EXECUTION]: The skill utilizes local shell commands, specifically git and the GitHub CLI (gh), to extract repository metadata, diffs, and history.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 21, 2026, 08:42 PM
Security Audit — agent-trust-hub — monthly-report-generator