fetch-diff

Pass

Audited by Gen Agent Trust Hub on Jun 22, 2026

Risk Level: SAFECOMMAND_EXECUTIONDATA_EXFILTRATIONPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The script scripts/fetch_diff.py executes the gh auth token command to retrieve authentication credentials if the GH_TOKEN environment variable is missing. This is a legitimate integration with the official GitHub CLI for authentication.
  • [DATA_EXFILTRATION]: The skill accesses a GitHub authentication token and transmits it to api.github.com to authorize API requests. This behavior targets a well-known service and is necessary for the skill's intended functionality.
  • [PROMPT_INJECTION]: The skill has an indirect prompt injection surface because it processes untrusted data from GitHub pull requests. 1. Ingestion points: Pull request descriptions and diff content are fetched from the GitHub API in scripts/fetch_diff.py. 2. Boundary markers: The output is formatted with line numbers but lacks explicit delimiters or instructions to the agent to ignore embedded commands. 3. Capability inventory: The skill can execute the gh CLI and perform network operations via aiohttp. 4. Sanitization: The script parses the diff for formatting but does not perform safety-specific sanitization of the content before presenting it to the agent.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 22, 2026, 08:50 AM
Security Audit — agent-trust-hub — fetch-diff