summarize

Pass

Audited by Gen Agent Trust Hub on May 18, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill references an external CLI utility ('summarize') recommended for installation via Homebrew from a developer repository.
  • [COMMAND_EXECUTION]: The core logic involves executing shell commands using the 'summarize' binary to process user-input URLs or file system paths.
  • [CREDENTIALS_UNSAFE]: Documentation provides standard instructions for configuring API keys using local environment variables, avoiding hardcoded secrets.
  • [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection due to its core function of processing untrusted data.
  • Ingestion points: Content is fetched and processed from arbitrary URLs and file paths provided as arguments to the 'summarize' command in SKILL.md.
  • Boundary markers: No explicit boundary markers or 'ignore' instructions are used to delimit the summarized content from agent instructions.
  • Capability inventory: The skill performs shell command execution using the 'summarize' binary.
  • Sanitization: No sanitization or filtering of the processed content is specified in the instruction set.
Audit Metadata
Risk Level
SAFE
Analyzed
May 18, 2026, 01:02 AM
Security Audit — agent-trust-hub — summarize