codex-history-ingest

Extracts knowledge from Codex CLI session logs into an Obsidian wiki.

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

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve? Past Codex CLI sessions contain valuable decisions, patterns, and solutions buried in noisy JSONL rollout logs. This Skill mines that history and distills durable knowledge into a structured Obsidian wiki instead of letting insights stay locked in transcripts. ## Core Features & Use Cases - Incremental Ingestion: Compares ~/.codex session files against a vault manifest to process only new or modified rollouts, with a full re-ingest mode available. - Structured Parsing: Reads session_index.jsonl and rollout-*.jsonl event envelopes, filtering telemetry noise and redacting secrets before extraction. - Topic Clustering & Distillation: Groups sessions by theme and writes wiki pages with provenance markers, confidence scores, and lifecycle metadata. - Use Case: After months of coding with Codex, run an ingest to surface recurring debugging patterns and project decisions as searchable wiki pages. ## Quick Start Ask the assistant to process your Codex history and add the extracted knowledge to your Obsidian wiki.

Frequently Asked Questions about codex-history-ingest

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

FAQPage Schema
How do I import my Codex CLI history into Obsidian?

Run the codex-history-ingest skill, which scans ~/.codex for session rollouts and index files, then distills them into wiki pages. It resolves your vault path from config and updates the manifest, index, and log after processing.

What files does Codex store session history in?

Codex stores sessions under ~/.codex/sessions/YYYY/MM/DD as rollout-*.jsonl event streams, plus a session_index.jsonl inventory and optional history.jsonl transcript. The skill prioritizes the index and rollout files over SQLite runtime databases.

Does the ingest process handle API keys or secrets in transcripts?

Yes, the skill applies a critical privacy filter that removes API keys, tokens, passwords, and credentials before writing anything to the wiki. It summarizes content instead of quoting raw transcripts and redacts private identifiers by default.

How does incremental sync avoid reprocessing old sessions?

Append mode compares each source file against .manifest.json in the vault root, processing only files that are new or modified since their recorded ingested_at timestamp. Full mode reprocesses everything when explicitly requested.

Will it create one wiki page per Codex session?

No, the skill clusters content by stable topics across sessions rather than mapping pages to sessions. It routes distilled knowledge into projects, concepts, skills, entities, or synthesis folders following existing wiki conventions.