lookbook-designer

Pass

Audited by Gen Agent Trust Hub on Aug 5, 2026

Risk Level: SAFEPROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
  • [PROMPT_INJECTION]: Indirect Prompt Injection Surface\n
  • Ingestion points: The skill reads project context and inspiration analysis using the 'renku inspiration analysis show' command as documented in 'SKILL.md' and 'references/using-inspiration-sources.md'.\n
  • Boundary markers: No explicit delimiters or instructions are used to distinguish ingested data from the agent's core instructions.\n
  • Capability inventory: The agent can execute CLI commands that modify project state, including 'renku lookbook apply' and 'renku media import', and uses shell commands like 'cd' and 'find'.\n
  • Sanitization: Input data from inspiration analyses is not explicitly sanitized or validated before being synthesized into JSON lookbook documents.\n- [COMMAND_EXECUTION]: Project CLI and Shell Command Integration\n
  • The skill instructs the agent to use the 'renku' CLI for project management and lookbook operations.\n
  • The agent is instructed to use shell commands 'cd' and 'find' to inspect project inspiration folders based on paths provided by the CLI tool.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 5, 2026, 02:13 AM
Security Audit — agent-trust-hub — lookbook-designer