claw-insights and codesession-cli

Claw-insights offers agent observability through session replay and metrics, while codesession-cli complements it by providing cost tracking for AI agent workflows, specifically integrating with OpenClaw (likely referring to the Claw-insights ecosystem).

claw-insights
54
Established
codesession-cli
50
Established
Maintenance 13/25
Adoption 11/25
Maturity 18/25
Community 12/25
Maintenance 13/25
Adoption 10/25
Maturity 18/25
Community 9/25
Stars: 5
Forks: 1
Downloads: 901
Commits (30d): 0
Language: TypeScript
License: MIT
Stars: 7
Forks: 1
Downloads: 522
Commits (30d): 0
Language: TypeScript
License: MIT
No risk flags
No risk flags

About claw-insights

LucaL6/claw-insights

Open-source agent observability — session replay, metrics, and shareable snapshots for AI agent workflows

Operates as a zero-intrusion read-only sidecar that tails logs and CLI output without SDK integration, storing data locally in SQLite and exposing observability through GraphQL APIs and Server-Sent Events. Built with an adapter architecture for extensibility beyond OpenClaw, it includes structured AI agent skills for autonomous setup and snapshot capture via REST API or CLI. Tech stack spans Express, GraphQL Yoga, SQLite, Satori rendering, and a React 19 frontend with real-time ECharts dashboards for token analytics and error tracking.

About codesession-cli

brian-mwirigi/codesession-cli

Track what your AI agents cost including files, commits, tokens, budgets. CLI , Dashboard+ API. Works with OpenClaw, Claude Code, and any agent framework.

Provides local API interception via an HTTP proxy (listening on localhost only) that captures Anthropic and OpenAI API calls in real-time, storing all metadata in a SQLite database without persisting prompts or keys. Integrates via ClawHub skills, Model Context Protocol, or wraps any script with `cs run`, plus offers a local web dashboard with budget alerts, session timelines, and per-model cost breakdowns—all without cloud dependency or telemetry.

Scores updated daily from GitHub, PyPI, and npm data. How scores work