AGENT LAB: SKILLS

agentation-self-driving

Pass

Audited by Gen Agent Trust Hub on Feb 17, 2026

Risk Level: SAFEPROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
  • [PROMPT_INJECTION] (LOW): Indirect Prompt Injection Surface.\n
  • Ingestion points: The agentation_watch_annotations tool in references/two-session-workflow.md ingests data that originates from external web pages scanned by the Critic agent.\n
  • Boundary markers: No specific delimiters or instruction-ignore warnings are defined to separate untrusted web content from the agent's instructions.\n
  • Capability inventory: The Fixer agent has the capability to read local files and perform file-write operations to the codebase.\n
  • Sanitization: No sanitization or escaping mechanisms are described for the annotation text before it influences code edits.\n- [COMMAND_EXECUTION] (SAFE): Intended Code Modification. The workflow grants an agent the ability to 'make the fix' by editing the local codebase. While this is a high-privilege capability, it is the stated primary purpose of the skill and is considered safe in that context, provided inputs are trusted.
Audit Metadata
Risk Level
SAFE
Analyzed
Feb 17, 2026, 04:57 PM