Parse Pi agent session JSONL files into readable conversation histories and diagnostic reports.
npx skills add https://github.com/OrestesK/pi --skill session-reader-orestesk
Please help me install this Agent Skill. Skill: session-reader Source: https://github.com/OrestesK/pi/tree/main/skills/session-reader Command: npx skills add https://github.com/OrestesK/pi --skill session-reader-orestesk
This Skill transforms dense Pi agent session JSONL files into readable, navigable conversation histories, making it easier to understand what happened, diagnose failures, and review agent activity.
Use the session-reader skill to analyze the attached Pi session JSONL file and summarize what happened, including any errors and tool usage.
High-intent search queries and answers about installing and using this skill.
You can parse Pi agent session JSONL files into readable conversation histories by running the session-reader script, which transforms dense logs into navigable text and diagnostic reports.
The best way to debug unexpected Pi agent results is to generate a table of contents from the session JSONL, locate the relevant exchange, and inspect the detailed turn for tool call errors or subagent activity.
Yes, you can analyze token cost and subagent activity by applying the dedicated cost analysis and subagent investigation modes to the Pi session JSONL file.
Parsing Pi session JSONL files requires Python 3.12 or newer and uses only standard-library dependencies, meaning no external packages or installations are needed.
Available analysis modes for Pi agent session JSONL data include conversation, table of contents, overview, detailed turn, issues, tools, costs, subagents, and full-session views.
You can search for specific keywords or truncate content during debugging by applying targeted investigation parameters, which support keyword search, content truncation, turn ranges, and error detection.
TaylorHustonEnd active learning sessions by generating summaries and updating session logs.
sky770825Analyze Syslog, JSON, and plain text logs to extract errors and generate reports.
Tyler-R-KendrickParse local JSONL session logs to report token usage and session savings.
papachongSearch JSONL session logs for historical chat messages with jq and ripgrep.
