skills/jellydn/my-ai-tools/pickup/Gen Agent Trust Hub

pickup

Pass

Audited by Gen Agent Trust Hub on Oct 8, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill is designed to read and process instructions from external Markdown files located in the .planning/handoffs/ directory. 1. Ingestion points: Local files in .planning/handoffs/ mentioned in SKILL.md. 2. Boundary markers: The skill instructions do not provide delimiters or warnings to ignore potentially malicious embedded commands in the handoff files. 3. Capability inventory: The agent can list files via shell (ls), read local files, and is directed to proceed with steps described in the handoff content. 4. Sanitization: No sanitization or validation of the handoff file content is specified.
  • [COMMAND_EXECUTION]: The skill uses the ls -la command to browse the filesystem for available handoff sessions.
Audit Metadata
Risk Level
SAFE
Analyzed
Oct 8, 2026, 07:44 AM