cognee-switch-datasets

Pass

Audited by Gen Agent Trust Hub on Sep 8, 2026

Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill instructs the agent to execute shell commands to run Python scripts (switch-dataset.py and sync-session-to-graph.py) located in the plugin's root script directory.
  • [INDIRECT_PROMPT_INJECTION]: The skill processes dataset names from script output and user arguments, using them in shell commands without defined sanitization steps.
    1. Ingestion points: Dataset names are ingested from the JSON output of switch-dataset.py --list --json as specified in SKILL.md.
    1. Boundary markers: There are no instructions to use delimiters or ignore instructions that might be embedded in the dataset names.
    1. Capability inventory: The agent is permitted to execute Python scripts via shell commands as seen in SKILL.md.
    1. Sanitization: The instructions do not specify any validation or escaping for the dataset names before they are interpolated into shell commands.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 8, 2026, 12:59 AM
Security Audit — agent-trust-hub — cognee-switch-datasets