summarize
Pass
Audited by Gen Agent Trust Hub on Jul 13, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill metadata specifies the installation of a binary from a third-party Homebrew tap (
steipete/tap/summarize) maintained by the author. - [COMMAND_EXECUTION]: The skill requires the agent to execute the
summarizecommand-line utility to perform its functions, which includes passing user-provided file paths and URLs as arguments. - [PROMPT_INJECTION]: The skill processes untrusted external data (URLs, YouTube transcripts, and local PDFs), creating a surface for indirect prompt injection attacks.
- Ingestion points: The
summarizecommand accepts URLs, YouTube links, and local file paths (e.g.,/path/to/file.pdf) as input for processing (SKILL.md). - Boundary markers: Absent. The skill instructions do not provide delimiters or specific warnings for the agent to ignore instructions embedded within the processed content.
- Capability inventory: The skill utilizes shell command execution via the
summarizeCLI tool (SKILL.md). - Sanitization: Absent. There is no evidence of content sanitization or validation before the ingested data is passed to the underlying model for summarization.
Audit Metadata