hermes-history-ingest

Extracts knowledge from Hermes agent memories and session logs into an Obsidian wiki.

Updated Nov 9, 2023
One-click install
npx skills add https://github.com/oresttokovenko/dot-files --skill hermes-history-ingest-oresttokovenko
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: hermes-history-ingest
Source: https://github.com/oresttokovenko/dot-files/tree/main/chezmoi/dot_agents/skills/hermes-history-ingest
Command: npx skills add https://github.com/oresttokovenko/dot-files --skill hermes-history-ingest-oresttokovenko

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve? Your Hermes agent accumulates valuable knowledge in ~/.hermes memories and session transcripts, but it stays locked in raw files instead of becoming searchable, organized documentation in your Obsidian vault. ## Core Features & Use Cases - Memory and Session Parsing: Reads Markdown/JSON memories and JSONL session transcripts, prioritizing high-signal content over tool-call noise. - Incremental Sync: Uses a .manifest.json delta check so only new or modified files are re-ingested, with a full mode for rebuilds. - Privacy-Aware Distillation: Redacts credentials and sensitive data, summarizes rather than quoting transcripts, and clusters knowledge by topic into wiki pages with provenance markers. - Use Case: After weeks of Hermes sessions on a project, run the ingest to distill recurring patterns, decisions, and techniques into projects/, concepts/, and skills/ pages in your wiki. ## Quick Start Process my Hermes history and add the extracted knowledge to my Obsidian wiki.

Frequently Asked Questions about hermes-history-ingest

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I import my Hermes agent history into Obsidian?

Run the hermes-history-ingest skill, which scans ~/.hermes memories and sessions, extracts durable knowledge, and writes distilled pages into your Obsidian vault. It resolves the vault path from config and updates the manifest, index, and log automatically.

What data does Hermes store in the ~/.hermes folder?

Hermes stores persistent memories as Markdown or JSON files, session transcripts as JSONL files organized by date, and a config.yaml. The .hub directory contains Skills Hub state and is skipped during ingest.

Does the ingest process protect sensitive information in session logs?

Yes. The skill redacts API keys, tokens, passwords, and credentials, summarizes rather than quoting raw transcripts, and asks before storing personal details. Privacy filtering is a mandatory step before any content reaches the wiki.

Will re-running the ingest duplicate existing wiki pages?

No. Append mode checks .manifest.json and only processes new or modified files since the last ingest. Full mode reprocesses everything but merges knowledge into existing pages rather than creating duplicates.

How are Hermes memories organized in the wiki after ingest?

Memories are clustered by topic, not one page per memory. Content routes to projects/, concepts/, skills/, entities/, or synthesis/ folders based on type, with provenance markers and confidence metadata on each page.