A Bugsnag MCP server for interacting with Bugsnag API
Nothing to run from here
Tracehub is installed from its publisher's own source and answers where it runs, so this marketplace is not in the path of a single call. There is no address here to send one to, and a panel that pretended otherwise would be showing you an answer we made up. Install it and call it from your own client — the Installation tab has the entry for each one.
99/100
trust score
17
downloads a day
local
runs on your machine
394
downloads in 30d
What it does
TraceHub MCP lets AI agents query and analyze OpenTelemetry traces from LLM applications. It parses gen_ai.* / llm.* semantic-convention attributes to enable automated debugging, cost analysis, and observability across Jaeger, Grafana Tempo, Traceloop, Datadog, Sentry, AWS X-Ray, New Relic, and Honeycomb backends. 19 read-only MCP tools: trace/span search, triage_trace root-cause diagnosis, correlate_trace across two backends, token/cost usage, model stats, sessions, prompt versions, and cost/error-spike investigation.
Quickstart
# 1 — add it to Claude Code
claude mcp add --scope user --env 'BACKEND_TYPE=${BACKEND_TYPE}' --env 'BACKEND_URL=${BACKEND_URL}' --transport stdio tracehub-mcp '--' uvx tracehub-mcp
# 2 — ask your agent for something these tools do
# search_traces, get_trace, triage_trace and 16 more
It needs BACKEND_TYPE, BACKEND_URL: this reads them from your environment, so export them first.
Tracehub costs nothing on mcprush: there is no plan to choose here, no cap to set and nothing mcprush can bill you for.
Publishernot claimed
Installation
Where are you running it?
Every route below installs the same thing and ends at the same approval screen. This one runs on your machine: your client starts Tracehub as a process under your own user, with your files and your network, so the tool surface below is what it can do to your computer rather than to a server somewhere else. Its scan result is on the Security tab — read the surface before you approve it.
Claude Codedeclared by this listing
This is a public server: you run it yourself and this marketplace is not in the path. You need uv (docs.astral.sh/uv): uvx comes with it. Claude Code registers it in one command; --scope user makes it available in every project. Claude Code reads BACKEND_TYPE (Which observability backend to query), BACKEND_URL (Backend API endpoint, e.g. http://localhost:16686 for Jaeger or https://api.datadoghq.com for Datadog) from your environment each time it starts the server, so the value stays out of its config: export them in your shell profile (macOS, Linux), or run setx BACKEND_TYPE …, setx BACKEND_URL … and open a new terminal (Windows). The '--' is in quotes so that PowerShell passes it on.
claude mcp add --scope user --env 'BACKEND_TYPE=${BACKEND_TYPE}' --env 'BACKEND_URL=${BACKEND_URL}' --transport stdio tracehub-mcp '--' uvx tracehub-mcpReconnect, or start a new session, and the tools appear in the model’s tool list.
Free on mcprush, and it runs from its publisher’s own package: this marketplace is not in the path, so there is no key of ours to hold and no card to add.
What it asks of you
It reaches a system of yours, so it needs settings only you can give it: BACKEND_TYPE — Which observability backend to query.; BACKEND_URL — Backend API endpoint, e.g. http://localhost:16686 for Jaeger or https://api.datadoghq.com for Datadog.. You set them in your own client’s config, on your machine — this marketplace never holds it.
What it writes
One config entry your client uses to start the process locally. A local server runs with your file system and your network, which is why it is priced without metering.
What you approve
19 tools, with what each one reads, writes and reaches shown before you agree — the same list on every route above. Read the tool surface.
Tool surface
What the model actually sees. Descriptions are diffed on every release — see version history.
No parameter schema on file.
Get complete trace details by trace ID.
No parameter schema on file.
Synthesize a likely-root-cause diagnosis for a trace, instead of returning raw trace data for the caller to re-derive one from every time.
No parameter schema on file.
Try to find the corresponding trace in a second, independently- configured backend (e.g. a Datadog trace and its downstream Sentry error, joined) - given a trace_id known to the primary backend.
No parameter schema on file.
Get aggregated LLM usage metrics (token counts) for a time period.
No parameter schema on file.
List all available services in the OpenTelemetry backend.
No parameter schema on file.
Find traces with errors.
No parameter schema on file.
List all LLM models being used with usage statistics.
No parameter schema on file.
Get detailed performance statistics for a specific LLM model.
No parameter schema on file.
List conversations/sessions grouped by gen_ai.conversation.id.
No parameter schema on file.
Get detailed statistics for a single conversation/session.
No parameter schema on file.
Compare aggregated LLM usage metrics between two time windows.
No parameter schema on file.
Investigate an LLM cost spike: compare a recent window against a baseline and rank which models/services contributed most to the change.
No parameter schema on file.
Investigate an error-rate spike: compare a recent window against a baseline and rank which services/models/error types contributed most.
No parameter schema on file.
Get aggregated performance stats grouped by prompt name and version.
No parameter schema on file.
Find traces with highest LLM token usage.
No parameter schema on file.
Find slowest LLM traces by duration.
No parameter schema on file.
Search for individual OpenTelemetry spans with optional filters.
No parameter schema on file.
List all LLM tools being used by identifying traceloop.span.kind == tool.
No parameter schema on file.
Free
$0 forever
- Every tool, no call limit
- No card, and mcprush charges nothing for it
- Source published under Apache-2.0
- Runs on your machine — nothing of it reaches our gateway
- Nothing to cap, because mcprush bills nothing
What counts against your monthly calls
| Tool | Unit | Calls used | Out of the allowance |
|---|
Nothing here is billable: mcprush charges nothing to install Tracehub or to call it, at any volume.
Call allowance · this installenforced at the gateway
Monthly allowance, in calls
—
—
Alert at 80% · then the tool returns a normal MCP error
Used this month: —
ATrust score 99 / 100
Release 0.12.3 · scanned 10 Oct 2026
Two independent axes, because powerful and malicious are different questions. The grade is threat only. The capability level is blast radius, and it is never a penalty on the grade — it is priced as one subtract-only term in the score, where you can see it.
Threat score100
Client score99
Findings1 threat · 0 capability
Capabilityminimal
Coveragesource
Verificationrepo
Surface digest6cdef5235b9146f94e4ac3e2c48c4358126c8b0e3cb1106be2d9caa44bbce475
Score vectorthreat 100 → client 99
| Term | Level | What it prices | Points |
|---|---|---|---|
| capability-exposure | minimal | capability blast radius (minimal) — client exposure if the model is manipulated | −0 |
| verification-discount | repo | publisher verification (public source) — no provenance, but the source is public and inspectable | −1 |
| coverage-honesty | source | inspection depth (source) — how much of the target the scan could see | −0 |
score = clamp(round(threatScore − E_cap − E_ver − E_cov), 0, 100). No model, no opinion, no network call at scoring time: the same methodology version against the same bytes always produces this number, and every line above is a field of the report rather than a summary of it.
What the scan could actually read
A grade is only as meaningful as its coverage, so the scanner publishes its own depth before it publishes its result.
Tool surfaceread
Implementation sourceread
Package metadataread
Live transportnot available
Tools were statically extracted from the published source (19 recovered), not enumerated from a running server. Tool-poisoning, Unicode-smuggling, capability and toxic-flow analysis ran on this inferred surface, but a mis-parsed registration could be missed or mis-attributed, so tool-derived findings are capped below “confirmed”. To grade the real runtime surface, scan the running server: --command "npx -y <package>".
Capability — what it could do if the model were manipulated
Tags derived from each tool’s schema and the implementation, not from what the tool calls itself. minimal is the level these add up to.
| Tool | Capability tags | Why the tag was assigned |
|---|---|---|
| search_traces | no tags | |
| get_trace | no tags | |
| triage_trace | no tags | |
| correlate_trace | no tags | |
| get_llm_usage | no tags | |
| list_services | no tags | |
| find_errors | no tags | |
| list_llm_models | no tags | |
| get_llm_model_stats | no tags | |
| list_sessions | no tags | |
| get_session_stats | no tags | |
| compare_time_windows | no tags | |
| investigate_cost_spike | no tags | |
| investigate_error_spike | no tags | |
| get_prompt_version_stats | no tags | |
| get_llm_expensive_traces | no tags | |
| get_llm_slow_traces | no tags | |
| search_spans_tool | no tags | |
| list_llm_tools_tool | no tags |
Toxic-flow graph
The lethal trifecta, checked as a graph rather than as a checklist: untrusted input, a sensitive source and an external sink have to meet before there is a path worth worrying about.
The public result for this release does not print the flow graph, so there is nothing to show here. That is not the same as "no paths were found": what the scan did read is above, under coverage.
Supply chain and provenance
Integrityfirst-seen
Surface digest6cdef5235b9146f9
Scanned2026-10-10
This is the first scan of this surface here, so there is nothing yet to compare it against.
Every result on this tab comes from one deterministic pass over the published package — offline, rule by rule, and auditable line by line above. Same methodology version, same bytes, same score.
Release history
Always latest — the install command names no version, so it installs the newest release when it runs; the current release on file is 0.12.3. Pin a release below and the command asks for exactly that one.
0.12.3
10 Oct 2026
latest- Maintenance release, no tool changes: the same 19 read-only tools for querying OpenTelemetry traces across Jaeger, Grafana Tempo, Traceloop, Datadog, Sentry, AWS X-Ray, New Relic and Honeycomb. Fixed - PyPI keywords now list all eight backends (Sentry, New Relic and Honeycomb were missing). - MCP Registry description shortened to fit the registry's 100-character limit. - Docker image rebuilt with urllib3 2.8.0 and PyJWT 2.15.1, clearing the October pip-audit advisories; Trivy reports no fixable HIGH or CRITICAL findings. - CI token permissions scoped per job; Trivy configuration aligned across workflows. Install: uvx tracehub-mcp (PyPI 0.12.3, signed provenance) or ghcr.io/mcpsmiths/tracehub-mcp:0.12.3.
0.12.2
20 Sep 2026
No release note was published with this version.
0.10.0
18 Sep 2026
No release note was published with this version.
0.6.0
15 Sep 2026
No release note was published with this version.
0.4.0
undated
- The version this listing was on when mcprush began following its releases. The registry was not asked when it shipped, so this row carries no date.
—
no reviews yet
Who can review
One review per account, from a signed-in account that does not publish this listing. Nothing else is asked — you can say what you think before you install it. Publishers can reply once per review, and a review can be edited or deleted by whoever wrote it.
Sign in to write one
Any signed-in account that does not publish Tracehub can review it, once — before installing it or after. A review from an account that has installed it is marked as one.
No reviews yet
Nobody has reviewed this listing. The rating on the card is the mean of the reviews written here and nothing else, so there is no rating until somebody writes the first — which takes a signed-in account that does not publish it, and nothing else.