cs-refactor-ff
Pass
Audited by Gen Agent Trust Hub on May 20, 2026
Risk Level: SAFECOMMAND_EXECUTION
Full Analysis
- [COMMAND_EXECUTION]: The skill instructs the agent to execute shell commands for searching code (
grep), managing version control state (git restore), and running project-specific test suites (unit tests, integration tests, and type checkers). These are standard operations for a development-oriented tool. - [INDIRECT_PROMPT_INJECTION]: The skill exhibits a potential surface for indirect prompt injection as it ingests and processes untrusted data from the local environment.
- Ingestion points: The agent reads local source code files and configuration files such as
.codestable/attention.md. - Boundary markers: Absent; the instructions do not specify the use of delimiters or 'ignore' instructions to separate data from system prompts.
- Capability inventory: The skill allows for local file system modification ('代码直接落盘') and shell execution for search and testing.
- Sanitization: Absent; no explicit sanitization or validation of the ingested file content is performed before processing.
Audit Metadata