clientccn2-project-editor

Pass

Audited by Gen Agent Trust Hub on Mar 7, 2026

Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill executes shell commands for essential development tasks, including running the test suite (npm test, npx jest), performing linting (npm run lint), and scanning the codebase for patterns using grep and find. It also uses node -e to execute a script that parses and validates JSON configuration files from the project directory.\n- [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection as it processes data from untrusted external files within the project. Mandatory evidence chain:\n
  • Ingestion points: Multiple files including source code (src/), JSON configurations (res/config/), and documentation (document/GameDesignDocument.md) are read and processed by commands like scan_client, check_client_consistency, and edit_client_idea.\n
  • Boundary markers: The skill instructions do not specify the use of delimiters or provide instructions to the agent to disregard commands or instructions that might be embedded within the project files.\n
  • Capability inventory: The agent has a broad set of capabilities, including executing shell commands, running Node.js code, and writing or editing files throughout the project structure.\n
  • Sanitization: No sanitization or filtering logic is mentioned for the content read from the project files before it is incorporated into the agent's reasoning or output generation.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 7, 2026, 05:01 PM