by rajudandigam
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..
# Add to your Claude Code skills
git clone https://github.com/rajudandigam/agent-inspectGuides for using ai agents skills like agent-inspect.
Last scanned: 7/2/2026
{
"issues": [
{
"type": "npm-audit",
"message": "@vitest/coverage-v8: Vulnerability found",
"severity": "critical"
},
{
"type": "npm-audit",
"message": "@vitest/mocker: Vulnerability found",
"severity": "medium"
},
{
"type": "npm-audit",
"message": "esbuild: esbuild enables any website to send any requests to the development server and read the response",
"severity": "medium"
},
{
"type": "npm-audit",
"message": "vite: Vite Vulnerable to Path Traversal in Optimized Deps `.map` Handling",
"severity": "high"
},
{
"type": "npm-audit",
"message": "vite-node: Vulnerability found",
"severity": "medium"
},
{
"type": "npm-audit",
"message": "vitest: Vulnerability found",
"severity": "critical"
}
],
"status": "FAILED",
"scannedAt": "2026-07-02T07:32:21.144Z",
"npmAuditRan": true,
"pipAuditRan": true,
"promptInjectionRan": true
}agent-inspect is an open-source ai agents skill for AI coding assistants such as Claude Code, Codex CLI, and ChatGPT, built by rajudandigam. 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. It has 294 GitHub stars.
agent-inspect failed SkillsLLM's automated security scan, which flagged one or more high-severity issues. Review the Security Report section carefully before using it.
Clone the repository with "git clone https://github.com/rajudandigam/agent-inspect" and add it to your Claude Code skills directory (see the Installation section above).
agent-inspect is primarily written in TypeScript. It is open-source under rajudandigam on GitHub, so you can review or fork the full source.
Yes. SkillsLLM lists many other AI Agents skills you can browse and compare side by side. Open the AI Agents category from the badge at the top of this page, or use the Related Skills and comparison links further down to weigh agent-inspect against similar tools.
No comments yet. Be the first to share your thoughts!
Requires a passing catalog security scan. Resolve the flagged issues and resubmit to enable featuring.
npm install agent-inspect
1. Capture one real run
2. Find the causal failure
3. Ask your coding agent to inspect it
4. Lock the fix with a contract
5. Attach the share-checked evidence
| Workflow | What you do |
|---|---|
| Debug one run | Capture/import → tree / timeline / report → first causal failure |
| Prevent one regression | TraceContract / check → suite / cohort → CI gate |
| Share one safe artifact | Redact → verify-safe → offline bundle → optional Studio review |
Coding-agent loop (optional): local read-only @agent-inspect/mcp-server — see CODING-AGENT-LOOP.md.
Review as a team (optional): workspace + optional SQLite index → customer-owned Studio Beta (no AgentInspect-hosted cloud). Portfolio tiers: POSITIONING-AND-PORTFOLIO.md.
Commands below match the packed quickstart. Replace <run-id> with a value from list.
npm install agent-inspect
npx agent-inspect init --yes
node examples/agent-inspect-demo.mjs
npx agent-inspect list --dir .agent-inspect
# After copying a run id from list:
npx agent-inspect report <run-id> --dir .agent-inspect
npx agent-inspect check <run-id> --dir .agent-inspect
npx agent-inspect bundle <run-id> --dir .agent-inspect --profile share
npx agent-inspect verify-safe <run-id> --dir .agent-inspect
npx agent-inspect bundle verify .agent-inspect/bundles/<run-id>
# Optional coding-agent loop (dry-run by default):
npx agent-inspect mcp configure --client cursor
init scaffolds config and a demo script; the demo writes the trace. Guide: First trace in 5 minutes · repo · Golden path · Coding-agent loop
Unique position: a local evidence debugger for TypeScript agents—not a hosted APM, eval host, or prompt registry. Complements LangSmith/Langfuse/Phoenix; owns the laptop → PR loop.
Proven mechanisms (shipped):
| Mechanism | What you get |
|---|---|
| Faithful execution trees | Nested steps, tools, LLMs, status, duration on disk |
| First causal failure | Conservative ordered engine for what failed first |
| Deterministic checks / TraceContract (Beta) | Trajectory expectations without an LLM judge |
| Share-checked Evidence v2 | bundle + bundle verify offline artifacts |
| Read-only MCP coding-agent loop (Preview) | mcp configure + @agent-inspect/mcp-server over local traces |
No fabricated adoption counts. External pilot evidence is still pending—see Pilot kit.
| Path | Use when | Start |
|---|---|---|
| Manual / observe | Custom nesting or object methods | Getting started |
| AI SDK | Vercel AI SDK generateText / streamText |
@agent-inspect/ai-sdk |
| OpenAI Agents | OpenAI Agents JS | @agent-inspect/openai-agents |
| LangChain | Callbacks / LangGraph-via-LangChain | @agent-inspect/langchain |
| Structured logs | Logs already emitted | Log-to-tree |
| Harness | Fixture runner for real projects | @agent-inspect/harness |
| CI reporters | Failed-test artifacts | vitest · jest |
| Standards files | OpenInference / OTLP JSON | Standards |
Blessed starters (no API keys): examples/starters
Understand — what / view / tree · timeline · report · diff · sessions / activity
Prevent regressions — deterministic checks · typed TraceContract (Beta) · suites · cohorts · CI gates · Vitest/Jest reporters (artifact reporters today; TraceContract matchers not shipped)
Share safely — redaction profiles · scan · verify-safe · offline bundles · CI artifacts
Scale locally — workspace · optional SQLite index (Beta) · observed outcomes · viewer / TUI / VS Code
Review as a team — customer-owned Studio Beta · explicit file / GitHub / HTTP ingest (disabled by default) · no maintainer cloud
Interoperate — read-only MCP server (Preview) · OpenInference-compatible / OTLP GenAI-aligned mapping with known-loss reporting
Support labels: SUPPORT-LEVELS.md · Network: NETWORK-BEHAVIOR.md
| Scenario | Start |
|---|---|
| Wrong tool / intentional failure | broken-agent-debugging |
| Coding-agent MCP debug loop | coding-agent-debug-loop · CODING-AGENT-LOOP |
| CI trajectory gate + redact | ci-eval-redact · CI artifacts |
| Safe incident handoff | Safe sharing |
| Multi-agent / session retry | Sessions & outcomes |
| Customer-owned team review | Self-hosting · Studio (optional) |
| Design partner trial | Pilot kit |
All blessed starters (no API keys): examples/starters
.agent-inspect/ (or AGENT_INSPECT_TRACE_DIR)Details: Safe sharing · Network behavior · Security
Current release: 6.12.1 (eighteen linked npm packages). Stable launch candidate; eight-week adoption checkpoint in progress; external pilot evidence pending. Persisted schema 1.0. Node