readwise-prune

Pass

Audited by Gen Agent Trust Hub on Mar 17, 2026

Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill relies on a CLI utility named 'readwise' to execute document management tasks.
  • [PROMPT_INJECTION]: The skill processes content from untrusted external documents which may contain malicious instructions (indirect prompt injection).
  • Ingestion points: Article summaries and document metadata are ingested through the 'readwise prune' command.
  • Boundary markers: No specific technical delimiters are used to isolate untrusted content from the agent's instructions.
  • Capability inventory: The skill has the capability to delete documents via the '--delete' flag.
  • Sanitization: The skill implements a human-in-the-loop review process as a mandatory safety measure before any destructive actions are taken.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 17, 2026, 02:35 AM