cognee-remember

Pass

Audited by Gen Agent Trust Hub on Sep 8, 2026

Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTIONDATA_EXFILTRATION
Full Analysis
  • [COMMAND_EXECUTION]: The skill invokes a local script (cognee-remember.sh) and a CLI tool (cognee-cli) via shell commands to process and store data.
  • [INDIRECT_PROMPT_INJECTION]: The skill accepts user-provided data via the $ARGUMENTS variable, which is then used in shell commands.
  • Ingestion points: Data entered by the user into the $ARGUMENTS variable in the SKILL.md instructions.
  • Boundary markers: None are defined to separate the data from the command instructions or to provide safety context to the agent.
  • Capability inventory: The skill can execute shell scripts and binary tools on the host system.
  • Sanitization: There is no description of input validation or sanitization for the $ARGUMENTS content.
  • [DATA_EXFILTRATION]: The skill transmits information to a Cognee server for storage in the knowledge graph. This is a core feature of the skill's functionality.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 8, 2026, 12:59 AM
Security Audit — agent-trust-hub — cognee-remember