rajudandigam

rajudandigam / agent-inspect

Public

Local execution trees for TypeScript AI agents. agent-inspect helps you understand what happened inside an AI agent run — locally. It turns manual steps, tool calls, LLM calls, structured logs, failures, durations, and run metadata into readable execution trees you can inspect from the terminal. It is built for TypeScript/Node.js developers..

www.npmjs.compackageagent-inspect agent-observability agentic-ai ai ai-agent ai-debugging
24
17
100% credibility
Found May 14, 2026 at 32 stars -- GitGems finds repos before they trend. Get early access to the next one.
Sign Up Free
AI Analysis
TypeScript
AI Summary

Local tool that creates readable execution trees from AI agent runs or structured logs for terminal inspection and debugging.

How It Works

1
🔍 Discover agent-inspect

You hear about a simple local tool that shows exactly what your AI helper did step by step, like a family tree of actions.

2
📦 Add it to your project

You easily include it in your coding setup with one quick command.

3
🎭 Trace your AI agent's work

You wrap your agent's main task, and it quietly records every plan, tool use, and thought into a neat tree.

4
đź“‹ List your sessions

You check a simple list of your recent agent runs to pick one.

5
🌳 View the execution tree

You see a clear tree view in your terminal showing steps, times, errors, and what went right or wrong.

6
Go deeper
⚖️
Compare two runs

Spot differences between good and bad agent behaviors side by side.

đź“ś
Parse existing logs

Turn your app's log files into the same helpful trees without changing code.

âś… Debug like a pro locally

You quickly understand and fix your AI agent, all without sending data anywhere.

Sign up to see the full architecture

5 more

Sign Up Free

Star Growth

See how this repo grew from 32 to 24 stars Sign Up Free
Repurpose This Repo

Repurpose is a Pro feature

Generate ready-to-use prompts for X threads, LinkedIn posts, blog posts, YouTube scripts, and more -- with full repo context baked in.

Unlock Repurpose
AI-Generated Review

What is agent-inspect?

agent-inspect builds local execution trees for TypeScript AI agents, turning manual steps, tool calls, LLM invocations, structured logs, failures, and metadata into terminal-readable trees. It solves the problem of flat console logs in complex agent runs—planning, branching, retries—by providing CLI inspection without cloud uploads or accounts. Built for Node.js/TS devs, it supports manual tracing APIs, log parsing, live tailing, and exports to Markdown/HTML/OpenInference.

Why is it gaining traction?

Unlike hosted tools like LangSmith or Phoenix, it's fully local-first with no vendor lock-in, letting you diff runs, tail logs in real-time, or export for PRs/postmortems before production observability. The CLI commands—list, view, clean, logs, tail, export, diff—offer instant value on existing JSON/log4js outputs, plus optional LangChain callbacks and TUI previews. It handles agent inspection requirements locally, from execution context to disclosure previews, without shipping data.

Who should use this?

TypeScript devs building production AI agents—like support bots, RAG pipelines, or multi-tool planners—who debug inner-loop runs before observability stacks. Teams using local GitHub Actions runners or Terraform local execution for agent tests, needing quick tree views of failures/retries. Node.js shops with structured logs wanting agent inspector meaning without full tracing rewrites.

Verdict

Solid local debugging layer for TS agent devs; try it alongside enterprise tools (24 stars shows early maturity, but docs/examples are comprehensive). Low 1.0% credibility score reflects newness—test thoroughly for your logs—but v1.0 stabilizes core APIs/CLI for real workflows.

(198 words)

Sign up to read the full AI review Sign Up Free

Similar repos coming soon.