actions-optimiser

Pass

Audited by Gen Agent Trust Hub on Jun 19, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: Retrieves GitHub Action workflow definitions from public repositories using the GitHub CLI (gh). This is a legitimate feature for analyzing community patterns.
  • [COMMAND_EXECUTION]: Executes standard shell commands (gh, jq, base64) to fetch, filter, and decode repository data. These operations are used as intended for repository interaction and data processing.
  • [PROMPT_INJECTION]: The skill processes untrusted external data (public workflow YAML files), which constitutes an indirect prompt injection attack surface. 1. Ingestion points: Public .github/workflows/*.yml files retrieved via the GitHub API. 2. Boundary markers: Absent. 3. Capability inventory: The skill modifies the user's local workflow files based on the analyzed patterns. 4. Sanitization: Absent; the agent processes raw decoded text from external sources.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 19, 2026, 07:33 AM
Security Audit — agent-trust-hub — actions-optimiser