learn-second-opinion

Pass

Audited by Gen Agent Trust Hub on Jul 3, 2026

Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill utilizes the Bash tool for core operations such as fetching wiki content via maestro wiki get and analyzing git diffs. These operations involve executing shell commands based on user-provided arguments.
  • [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection as it ingests and processes untrusted data from local files, wiki entries, and git diffs.
  • Ingestion points: Content from file paths, wiki IDs (type-slug), or git HEAD/staged diffs provided via user arguments (SKILL.md).
  • Boundary markers: The instructions do not define specific delimiters (e.g., XML tags or backticks) to wrap the ingested content when it is passed to the sub-agents for analysis.
  • Capability inventory: The orchestration agent has access to Bash, Write, Edit, and spawn_agents_on_csv. Spawned persona agents are explicitly instructed not to modify files or call spawn tools, which significantly mitigates the risk of an injection attack leading to unauthorized changes.
  • Sanitization: There is no explicit sanitization or filtering of the target content before analysis.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 3, 2026, 08:44 AM
Security Audit — agent-trust-hub — learn-second-opinion