claude-history-ingest
Pass
Audited by Gen Agent Trust Hub on May 19, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill's core functionality is to discover and read local data files from Claude Code storage paths (~/.claude and macOS Application Support), which is consistent with its stated purpose of ingesting history.
- [SAFE]: Data processing is focused on knowledge extraction and distillation into a local Obsidian vault. No network exfiltration or transmission of sensitive conversation data to external servers was detected.
- [SAFE]: Shell command usage is limited to standard file system discovery using find and Glob operations within designated local application paths to locate session data.
- [SAFE]: The skill includes explicit instructions for the agent to avoid secrets, API keys, and sensitive personal information during the extraction process, and to prioritize distillation over raw verbatim copying.
Audit Metadata