what-did-i-get-done

Pass

Audited by Gen Agent Trust Hub on Aug 22, 2026

Risk Level: SAFEPROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
  • [PROMPT_INJECTION]: The skill ingests commit messages from a local git repository, which represents an untrusted data source. This creates an indirect prompt injection surface where a malicious commit message could attempt to override the agent's summary instructions or formatting rules.
  • Ingestion points: Commit messages retrieved via git log (referenced in SKILL.md workflow).
  • Boundary markers: No specific delimiters or boundary markers are defined to isolate commit data from the agent's instructions.
  • Capability inventory: Execution of git commands and text synthesis.
  • Sanitization: The skill does not define any sanitization, filtering, or validation for the content of the commit messages before they are processed by the LLM.
  • [COMMAND_EXECUTION]: The skill requires the ability to execute git commands (e.g., git log) to fetch authored commits. This is a functional requirement for the skill's purpose but involves interacting with the underlying system's shell environment.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 22, 2026, 09:17 AM
Security Audit — agent-trust-hub — what-did-i-get-done