agentrace MCP server

by ura-tools·io.github.ura-tools/agentrace·v1.0.0

Structured observability for AI agents. Trace steps, decisions, and errors. MCP server + CLI viewer.

B72/100grade B

What users say

No reviews yet

Safety scan

B72/100

Adoption

New

0 stars29 downloads/wk

Reviews

Nobody has reviewed agentrace yet.

If you have run it, two minutes of your experience saves the next person an afternoon.

write = sends, deletes, buys or posts

Read from the package source without running it. The installed server may list more.

trace_decision
Log a decision point. Use this when the agent chooses between alternatives — captures the reasoning behind the choice.
trace_end
End a trace session. Call this when the task is complete (successfully or not).
trace_error
Log an error or unexpected state encountered during execution.
trace_list
List recent traces. Useful for reviewing past agent sessions.
trace_start
Start a new trace session. Call this at the beginning of a complex task to track what the agent does, why, and what happens.
trace_step
Log a step in the current trace. Use this for each meaningful action: tool call, file read, code change, API call, etc.
trace_view
View a complete trace with all its events.

scanner v0.1.9 · 2026-09-27 · same rubric, same numbers if you re-run it

no findings

Code scan8 source files scanned25/25
–Live reliabilityno gateway calls yet and no remote to proben/a
–Tool poisoningtools not inspected (local package is not executed); not countedn/a
Auth qualitylocal package, no credentials required12/15
Maintenancelast push 183 days ago4/15
Maintainer identityregistry namespace matches repository owner6/10

Overall 72/100. Components that don't apply are left out of the denominator. Any critical finding is an F.Rubric Appeal a finding JSON

Against every graded MCP server, from our own scans and uptime checks. Updated after each scan run.

Safety score: 72/100; higher than 38% of 48,255 servers
Tools: 7; the typical server has 7
Tools that change things: 0%; typical: 0%. Fewer write tools means less an agent can break
Downloads a week: 29; more than 16% of 12,191 servers with a package

What its tools let an agent do: 7 read

Runs npx -y @ura-dev/agentrace on your machine. Read the scan report first; the gateway never runs local packages.

claude mcp add agentrace -- npx -y @ura-dev/agentrace

Is agentrace MCP server safe?: Mostly: it is graded B (72/100). Read the agentrace safety report
How do I install agentrace?: It runs on your machine. Copy the Claude Code, Cursor, VS Code or Claude Desktop config from the install section.
Does agentrace need an API key?: Not as far as the registry entry and our scan can tell: no credentials are declared or required.
Is agentrace maintained?: The last commit was 221 days ago (2026-03-28). The latest release is v1.0.0.
How does agentrace compare with other MCP servers?: safety score 72/100 (higher than 38% of 48,255 servers); tools 7 (the typical server has 7); tools that change things 0% (typical: 0%. Fewer write tools means less an agent can break). Full comparison.