skill-health

Pass

Audited by Gen Agent Trust Hub on Nov 3, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTIONDYNAMIC_EXECUTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted data from external session traces that could contain malicious instructions.
  • Ingestion points: The skill queries the 'input' field from Braintrust logs (as specified in SKILL.md), which stores arbitrary user messages from historical sessions.
  • Boundary markers: The skill does not implement delimiters or explicit 'ignore' instructions when the agent reads or processes these log inputs.
  • Capability inventory: The skill has the power to write to the filesystem (modifying SKILL.md and creating references/ files) and execute git commands.
  • Sanitization: There is no evidence of data sanitization or validation performed on the log content before it is used to generate instruction updates.
  • [DYNAMIC_EXECUTION]: The skill's core functionality involves the runtime generation and modification of SKILL.md files.
  • These markdown files act as executable instructions for the agent platform. By programmatically rewriting its own logic based on external log data, the skill performs a form of dynamic execution that can be exploited if the source data is compromised.
Audit Metadata
Risk Level
SAFE
Analyzed
Nov 3, 2026, 05:47 AM