# Pulse: Agent Cognition Observability

> See how your coding agents actually think, not just what they cost.

**Status:** Source-available  
**Repository:** https://github.com/michaewahl/pulse

## Problem

We can see our AI API bills but not whether agents are thrashing, re-reading the same files, or degrading over a session.

## What it does

A local observability tool that reads the JSONL session transcripts coding agents like Claude Code already write to disk and turns them into a health dashboard. It computes six indicators: file-revisit rate (thrashing), turns-to-resolution, context/cache efficiency, whether the agent actually queries the project memory it built, a frustration index from user-side sentiment, and swarm detection separating orchestrator from sub-agent sessions. The point is to pinpoint where architecture is making the model thrash. It runs 100% locally: no cloud, no API keys, no telemetry.

## Measured results

- **6 metrics** agent-cognition health
- **100% local** no cloud, no API keys — Parses transcripts already on disk; no data leaves the machine.

## Stack

TypeScript, Node, Local JSONL transcript parsing, HTML dashboard

## Topics

AI, Agents, DevOps

---

Part of the proof-of-work portfolio of Michael Wahl — VP, Software Engineering & AI Enablement.

- Canonical project page: https://connect.michaelwahl.org/project/pulse-agent-observability
- Full professional profile: https://connect.michaelwahl.org/
- Machine-readable context: https://connect.michaelwahl.org/llms.txt