Pheebs

Pheebs is an open-source AI telemetry tool that measures how developers work with AI coding agents.

Pheebs is an open-source telemetry tool created by Eversynced to understand how developers work with AI coding agents and what models cost them. It sits quietly inside Claude Code, Cursor, and Codex via hooks, capturing lightweight interaction signals like session shapes, event types, durations, and counts without storing source code, file contents, or prompt text. The tool records session telemetry locally in JSONL format and optionally sends data to a self-hosted or managed backend via HTTP routes. It generates insights on model spend, verification rates, and tool adoption by analyzing session difficulty against the models that ran. Pheebs supports engineering teams by auditing AI usage, sizing work tasks, and measuring proficiency through repertoire and judgement signals.

Pheebs at a glance

Best for: Software engineer, Director
Runs on: Web, API, self-hosted, open source
Pricing: Free
Released: October 2026
Category: AI coding tool
Website: eversynced.com

Key features

Captures session interaction signals through hooks
Logs session data locally in JSONL format
Exports native OpenTelemetry metrics and logs
Classifies prompt intent without storing prompt text
Tracks model spend and identifies overpaid models
Measures verification coverage and pushback rates
Resolves developer identity and git repositories

Use it for

Auditing team model spend on AI coding agents
Measuring verification coverage for AI-written code
Tracking AI tool and skill adoption across teams
Analyzing prompt intent and session interaction shapes

Alternatives to Pheebs

AI Engineer Tools — Capture agent activity as structured local events without conversations.
OpenCode — Works with 75+ model providers, local models, and existing subscriptions
Aegis — It ensures AI coding agents work against verified baselines and proof.
jevgrep — Reduces coding agent costs by providing targeted, relevant repository context.
Freebuff — Access multiple coding AI models for free without subscriptions.
pstack — Improves AI coding agent outcomes with rigorous workflows.
Genkit — Provides unified APIs and local developer tools for AI application workflows.
Qoder — It features deep codebase awareness, autonomous coding agents, and cross-platform support.
Entelligence AI — Inspect agent sessions locally without digging through logs or sharing data.
AI Deploy Bench — Measures AI agent success, time, tokens, cost, and human steps for cloud deployments.
Magpie — It enables AI agents to use models not natively supported by their internal protocols.
Coucou — It provides open-source, real-time agent monitoring and approvals directly from the screen notch.

All Pheebs alternatives