skills/cniska/skills/plan/Gen Agent Trust Hub

plan

Pass

Audited by Gen Agent Trust Hub on Apr 13, 2026

Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill instructs the agent to use the GitHub CLI tool with user-provided input (gh issue view $ARGUMENTS). If the agent implementation does not properly sanitize these arguments, it could lead to command injection in environments where the agent has shell access.
  • [PROMPT_INJECTION]: The skill possesses an indirect prompt injection surface as it ingests and processes untrusted data from external sources to guide its decision-making and planning logic.
  • Ingestion points: Untrusted content is ingested from GitHub issues (via gh issue view) and from the project's source code files (via read the relevant code).
  • Boundary markers: Absent. The instructions provide no clear delimiters or guidance to the agent to treat fetched data as data rather than instructions.
  • Capability inventory: The skill allows the agent to read repository files and interact with the GitHub CLI tool.
  • Sanitization: Absent. There is no evidence of content validation or escaping before the external data is incorporated into the design dialogue.
Audit Metadata
Risk Level
SAFE
Analyzed
Apr 13, 2026, 02:37 PM
Security Audit — agent-trust-hub — plan