deep-research

Pass

Audited by Gen Agent Trust Hub on Sep 22, 2026

Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill utilizes shell commands to interact with the host system's clipboard. Evidence includes the use of pbcopy, xclip, wl-copy, and clip.exe within a shell pipe chain to transfer generated content to the clipboard.
  • [INDIRECT_PROMPT_INJECTION]: The skill ingests data from local project files and user input to construct a prompt, creating a surface for indirect instructions to influence agent behavior.
  • Ingestion points: Reads project configuration files including package.json, requirements.txt, and Cargo.toml as instructed in SKILL.md step 2.
  • Boundary markers: Absent. There are no instructions to the agent to use delimiters or ignore potential instructions embedded within the configuration files it reads.
  • Capability inventory: The skill possesses the capability to execute shell commands to modify the system clipboard.
  • Sanitization: Absent. Content from the external configuration files is interpolated directly into the prompt structure without escaping, validation, or filtering.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 22, 2026, 02:10 PM
Security Audit — agent-trust-hub — deep-research