summarize

Pass

Audited by Gen Agent Trust Hub on Sep 19, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill metadata specifies the installation of a CLI tool via a third-party Homebrew tap ('steipete/tap/summarize').
  • [COMMAND_EXECUTION]: The skill operates by executing the 'summarize' binary with user-provided arguments, flags, and local file paths.
  • [INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted data from the web, PDFs, and YouTube links which may contain malicious instructions designed to override agent behavior.
  • Ingestion points: Web URLs, YouTube links, and local file paths (e.g., .pdf, audio, images) processed by the 'summarize' command in 'SKILL.md'.
  • Boundary markers: None. The skill does not define specific delimiters or instructions to ignore embedded commands within the content being summarized.
  • Capability inventory: Execution of the 'summarize' CLI, access to environment variables (API keys), and reading of local files and configuration at '~/.summarize/config.json'.
  • Sanitization: None described. The skill relies entirely on the safety filters of the underlying AI models (OpenAI, Anthropic, Google, xAI).
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 19, 2026, 11:31 PM
Security Audit — agent-trust-hub — summarize