html-debrief

Pass

Audited by Gen Agent Trust Hub on Aug 10, 2026

Risk Level: SAFE
Full Analysis
  • [COMMAND_EXECUTION]: The skill executes a local shell script (scripts/prepare.sh) to validate directory structures and slugs. This is a standard initialization step for the skill's functionality.
  • [DATA_EXPOSURE]: The skill creates and writes to a specific local directory (./.ai/debriefs/) within the current workspace. It implements collision checks to prevent accidental overwriting of files.
  • [EXTERNAL_DOWNLOADS]: The skill refers to a legitimate external resource (a GitHub repository by the same author, paulrberg) in an error message for installation instructions if a dependency is missing. This is informative and does not perform automatic downloads.
  • [PROMPT_INJECTION]: No prompt injection or bypass patterns were detected. The instructions focus on grounding report content in task evidence.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 10, 2026, 04:27 PM
Security Audit — agent-trust-hub — html-debrief