reflect
Pass
Audited by Gen Agent Trust Hub on May 8, 2026
Risk Level: SAFECOMMAND_EXECUTION
Full Analysis
- [COMMAND_EXECUTION]: The skill uses SQL queries to read from a local SQLite database (
~/.claude/ccrecall.db) usingmcp-sqlite-tools. This is used to retrieve conversation history for the purpose of identifying user corrections and learnings. - [COMMAND_EXECUTION]: The skill is designed to modify or create files in
.claude/skills/or~/.claude/skills/. The instructions explicitly state that the agent must show suggested updates and wait for user confirmation ('Approve - Wait for user confirmation before writing') before performing these writes.
Audit Metadata