skills/dafang/codestable/cs-note/Gen Agent Trust Hub

cs-note

Pass

Audited by Gen Agent Trust Hub on Jul 22, 2026

Risk Level: SAFEPROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
  • [PROMPT_INJECTION]: The skill manages the .codestable/attention.md file, which is designated as mandatory context for all other agent sessions, creating a surface for Indirect Prompt Injection.
  • Ingestion points: The skill reads existing content from .codestable/attention.md and accepts arbitrary text input from the user to append to the file.
  • Boundary markers: It uses a structural HTML comment <!-- cs-note managed --> to locate its management section but does not apply delimiters, escaping, or "ignore instructions" warnings to the individual facts provided by the user.
  • Capability inventory: The skill utilizes file read/write operations and the grep utility to modify project-wide configuration.
  • Sanitization: There is no validation or sanitization of user-provided content before it is written to the persistent project context.
  • [COMMAND_EXECUTION]: The instructions explicitly direct the agent to use the grep command-line tool to perform duplicate checks and search for keywords within the project's local files.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 22, 2026, 12:38 PM
Security Audit — agent-trust-hub — cs-note