experiment-protocol

Pass

Audited by Gen Agent Trust Hub on Mar 29, 2026

Risk Level: SAFEPROMPT_INJECTION
Full Analysis
  • [PROMPT_INJECTION]: The skill defines a workflow that ingests potentially untrusted data into the agent's context through experiment artefacts and user interactions.
  • Ingestion points: External files (artefacts) and user-supplied input collected during phase 1 and 2 (SKILL.md).
  • Boundary markers: The instructions do not specify any delimiters or ignore-instructions markers for the data being processed.
  • Capability inventory: The skill utilizes tools for file manipulation via the MCP server and triggers a handoff to the '@retrospective-analyst' for artefact processing (SKILL.md).
  • Sanitization: No sanitization or validation logic is defined for the content of the artefacts before they are used by the analyst agent.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 29, 2026, 08:42 AM