pulling-updates-from-skills-repository

Pass

Audited by Gen Agent Trust Hub on Sep 9, 2026

Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill instructs the agent to execute various shell commands using git to manage local repository state, including status, fetch, merge, and log in SKILL.md.
  • [COMMAND_EXECUTION]: The workflow concludes by executing a local script located at a relative path ../using-skills/find-skills to verify update success as specified in Step 8.
  • [INDIRECT_PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection because it reads and processes data from an external, potentially untrusted source (upstream Git repository).
  • Ingestion points: Data enters the context via git log HEAD..@{u} in Step 4 and through manual inspection of conflicted files in Step 6.
  • Boundary markers: Absent. There are no instructions to the agent to treat the remote commit messages or diffs as untrusted or to ignore instructions embedded within them.
  • Capability inventory: The skill has the ability to execute shell commands including git and the find-skills script, and modify the local file system.
  • Sanitization: Absent. The skill does not perform any validation, filtering, or escaping of the content retrieved from the remote repository before presenting it to the agent or user.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 9, 2026, 03:52 AM
Security Audit — agent-trust-hub — pulling-updates-from-skills-repository