pi-history-ingest

Ingest Pi session history into an Obsidian wiki while redacting sensitive data.

Updated Feb 2, 2016
One-click install
npx skills add https://github.com/zjykzk/blog --skill pi-history-ingest
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: pi-history-ingest
Source: https://github.com/zjykzk/blog/tree/main/.agents/skills/pi-history-ingest
Command: npx skills add https://github.com/zjykzk/blog --skill pi-history-ingest

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Ingest Pi coding agent session history into an Obsidian wiki to extract durable knowledge from past Pi sessions while maintaining privacy and constructive context for future work.

Core Features & Use Cases

  • Ingest and summarize sessions stored under ~/.pi/agent/sessions into topic-oriented wiki pages, enabling knowledge reuse.
  • Supports append-mode incremental updates and full re-ingest to ensure consistency with the vault and manifest state.
  • Extracts actionable insights from user and assistant messages, session_info titles, and compaction/branch summaries to surface patterns like project scope and workflow.

Quick Start

Trigger the wiki history ingest workflow against your vault using the configured PI_HISTORY_PATH and OBSIDIAN_VAULT_PATH.

Frequently Asked Questions about pi-history-ingest

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

FAQPage Schema
How do I extract durable knowledge from Pi coding agent sessions?

Ingesting Pi sessions extracts durable knowledge by parsing local session stores, such as ~/.pi/agent/sessions, and organizing user and assistant messages into topic-oriented Obsidian wiki pages for knowledge reuse.

Can I automatically redact sensitive data when importing Pi session history?

Yes, you can redact sensitive data during Pi session history ingestion. The workflow automatically applies privacy redaction to confidential details found in tool results and compaction summaries before writing to the wiki.

Does the Obsidian wiki ingestion support incremental updates for new Pi sessions?

Obsidian wiki ingestion supports incremental updates through an append-mode feature, allowing you to process new or updated Pi sessions and maintain consistency with the vault and manifest state without a full re-ingest.

What is the best way to organize Pi session compaction summaries in a wiki?

Organizing Pi session compaction summaries in a wiki is best handled by extracting insights into topic-oriented pages. The ingestion process surfaces workflow patterns and project scope from compaction and branch summaries.

Do I need a specific local directory structure to ingest Pi sessions?

You need a local Pi session store, typically located at ~/.pi/agent/sessions, and a configured Obsidian vault path. The ingestion workflow processes session JSONL files from these directories to build your knowledge base.

Why are my Pi session tool results not showing up in the Obsidian wiki?

Pi session tool results might be missing if they contain confidential details that triggered privacy redaction rules, or if the ingestion extraction rules failed to parse the session JSONL format correctly during the wiki update.