ReasoningBank with AgentDB
Pass
Audited by Gen Agent Trust Hub on Mar 26, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill uses
npx agentdb@latestto initialize and manage the database. This fetches theagentdbpackage from the npm registry.\n- [COMMAND_EXECUTION]: The skill instructions contain shell commands for database initialization (init), migration (migrate), and Model Context Protocol (MCP) server setup. These operations manage local files like.agentdb/reasoningbank.db.\n- [PROMPT_INJECTION]: The skill processes agent trajectories and experiences that may contain untrusted content (Indirect Prompt Injection surface). \n - Ingestion points: Data enters via
rb.insertPatternandnpx agentdb importinSKILL.md.\n - Boundary markers: No specific delimiters or instructions to ignore embedded commands are present in the examples.\n
- Capability inventory: The skill interacts with the local file system and executes database commands.\n
- Sanitization: Content is stored as JSON but processed via
synthesizeContextwhich may interpolate data into prompts.
Audit Metadata