recall

Pass

Audited by Gen Agent Trust Hub on Apr 18, 2026

Risk Level: SAFECOMMAND_EXECUTIONDATA_EXFILTRATIONPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill uses find and grep shell commands to perform keyword searches within local conversation history files.
  • [DATA_EXFILTRATION]: The skill reads from sensitive application data files, including history.jsonl and MEMORY.md, which store session transcripts and memories. This access is necessary for the recall feature and does not involve exfiltration to external domains.
  • [PROMPT_INJECTION]: The skill has an indirect prompt injection surface because it processes data from previous conversations which may contain untrusted content.
  • Ingestion points: Data is read from local history.jsonl and MEMORY.md files.
  • Boundary markers: There are no markers or instructions defined to isolate recalled content from the agent's command logic.
  • Capability inventory: Uses shell commands (find, grep) and memory tools.
  • Sanitization: No evidence of content sanitization exists for the recalled data.
Audit Metadata
Risk Level
SAFE
Analyzed
Apr 18, 2026, 12:08 PM
Security Audit — agent-trust-hub — recall