traceai operator
Operate, analyze, and troubleshoot TraceAI CLI for local-first agent observability and MCP proxying.
Free
Works with the AI Agents you already use
traceai operator
Example session with this skill installed
I need to audit my local tool performance. Build the TraceAI binary, verify health, and then show me the top 5 most frequently called tools from the default SQLite database.
- Read your context and instructions
- Compiled the traceai operator
I have built the TraceAI binary and verified the local environment. The system is healthy, and I have extracted the tool usage rankings from the local SQLite store.
Local Audit Results
Binary built at bin/traceai. Health check returned ok.
| Tool Name | Call Count | Success Rate | Avg Latency |
|---|---|---|---|
filesystem_read | 142 | 98.6% | 12ms |
web_search | 85 | 82.4% | 840ms |
calculate_metrics | 44 | 100% | 5ms |
git_commit | 31 | 90.3% | 45ms |
slack_notify | 12 | 58.3% | 1200ms |
Next steps
- Run
traceai high-failures --threshold 0.3to investigate the low success rate of theslack_notifytool. - Launch
traceai dashboardto visualize the call sequences for these tools over time. - Configure
traceai cleanup --retention-days 7if you want to rotate this audit data weekly.
Connects securely to your tools. The creator never sees your data.
About this skill
Development teams struggle to gain visibility into local tool calls, MCP interactions, and agentic workflows without complex cloud infrastructure. Local observability often feels like a black box, leading to silent tool failures or unoptimized agent behavior.
What it does
- CLI Management handles installation, local Go builds, and versioning for the TraceAI binary across platforms.
- MCP Proxying wraps existing Model Context Protocol servers to capture and log JSON-RPC traffic without breaking protocol communication.
- Local Telemetry manages SQLite-backed storage for tool events, function calls, and agent metrics with configurable retention.
- Analysis & Reporting generates rankings for tool usage, failure rates, call sequences, and retry patterns directly from the terminal.
- Health & Monitoring runs diagnostic checks, status reports, and JSON-based metrics smoke tests to verify collector integrity.
How it works
- Bootstrap the environment by installing via npm, Go, or building the
traceaibinary from source. - Instrument workflows by replacing standard MCP commands with the proxy or integrating the Go SDK.
- Analyze captured data using commands like
top-tools,high-failures, orcall-seqto identify bottlenecks. - Export results to CSV or JSON formats for external reporting or startup the local Dashboard for visual debugging.
Frameworks & tools
Built for Go 1.25+ and Node.js 18+. Integrates with the Model Context Protocol (MCP), SQLite for local storage, and provides HTTP/gRPC middleware support.
Why this beats prompting it yourself
Writing reliable CLI wrappers for telemetry requires strict handling of stdout/stderr to avoid polluting protocol channels. This skill encodes the specific flag ordering, database locking logic, and capture-mode privacy defaults needed to operate TraceAI safely without manual trial and error.
Use cases
- Debugging MCP server communication by proxying and logging local tool calls.
- Auditing agent performance to identify high-failure tools or zero-call redundancies.
- Setting up local observability dashboards for development environments without external dependencies.
- Automating release validation by running health checks and metric exports post-build.
Known limitations
Does not support long-running processes unless explicitly requested. Database cleanup is only effective if retention days or byte limits are pre-configured. Standard output is reserved for protocol data during proxying.
Requires local SQLite file permissions. Proxy mode requires strict stdout isolation. Database cleanup needs explicit retention configuration.
How to use
- 1
Connect your agent to the Agensi MCP
Copy the prompt and paste it into Claude, ChatGPT, Cursor or any agent that supports MCP. It sets up the connection for you. Manual setup
- 2
Sign in to Agensi
Your agent shows a sign-in link. Sign in and click Allow.
- 3
Ask for what you need
Your agent loads this skill and uses it on your task. It always asks before any payment.
Reviews
No reviews yet
Be one of the first to try it. Every listed skill goes through the trust checks below.
Security scanned
Checked by our 8-point scan before listing
1 use
Times used to date
Free forever
No account required to browse
Trust & safety
Security scan
Security scan passed 1 month ago
- Free with an Agensi account
Requires