give-me-tips

Pass

Audited by Gen Agent Trust Hub on Sep 19, 2026

Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill executes senpi --list-tips to obtain a live list of tips in JSON format, which it uses to identify the specific feature a user is asking about.
  • [INDIRECT_PROMPT_INJECTION]: The skill has an attack surface for indirect prompt injection as it ingests and processes external code and catalog data to generate its explanations.
  • Ingestion points: Retrieves data from the senpi --list-tips command output and reads source files from the @code-yeongyu/senpi and oh-my-openagent packages.
  • Boundary markers (absent): No specific delimiters are used to separate ingested content from instructions.
  • Capability inventory: The skill reads local files and executes the senpi CLI tool.
  • Sanitization (absent): The skill does not perform specific sanitization of the ingested code text, as its primary function is to summarize and explain technical logic to the user.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 19, 2026, 06:00 PM
Security Audit — agent-trust-hub — give-me-tips