Model Context Protocol server for Rollbar
Nothing to run from here
Downtrace 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.
100/100
trust score
9
downloads a day
local
runs on your machine
663
downloads in 30d
What it does
Ask a Downtrace project what changed in a Node.js backend in production, and why.
Quickstart
# 1 — add it to Claude Code
claude mcp add --scope user --env 'DOWNTRACE_URL=${DOWNTRACE_URL}' --transport stdio downtrace-mcp '--' npx -y @downtrace/mcp
# 2 — ask your agent for something these tools do
# read_credential, list_projects, project_status and 26 more
It needs DOWNTRACE_URL: this reads it from your environment, so export it first.
Downtrace costs nothing on mcprush: there is no plan to choose here, no cap to set and nothing mcprush can bill you for.
Publishernot claimed
RAAll 1 server
RadW20201 server
1 category
Collected from a public index. Nobody has claimed this account, so nothing here was written by its author — claim it if it is yours.
Installation
Where are you running it?
Every route below installs the same thing and ends at the same approval screen. This one runs on your machine: your client starts Downtrace as a process under your own user, with your files and your network, so the tool surface below is what it can do to your computer rather than to a server somewhere else. Its scan result is on the Security tab — read the surface before you approve it.
Claude Codedeclared by this listing
This is a public server: you run it yourself and this marketplace is not in the path. You need Node.js — the current LTS (24) is the safe choice; npx comes with it. Claude Code registers it in one command; --scope user makes it available in every project. Claude Code reads DOWNTRACE_URL (The address of your Downtrace) from your environment each time it starts the server, so the value stays out of its config: export it in your shell profile (macOS, Linux), or run setx DOWNTRACE_URL … and open a new terminal (Windows). The '--' is in quotes so that PowerShell passes it on.
claude mcp add --scope user --env 'DOWNTRACE_URL=${DOWNTRACE_URL}' --transport stdio downtrace-mcp '--' npx -y @downtrace/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: DOWNTRACE_URL — The address of your Downtrace. You set it 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
29 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.
Read from its source code. A server that builds its tool list when it starts can list more tools than this.
What the credential this server was given can do, asked before attempting anything: the project it belongs
No parameter schema on file.
The list of projects, in the order the front page reads them: the ones with open findings first, then by
No parameter schema on file.
What a project looks like right now: traffic, endpoints, dependencies, runtime health, coverage
| Parameter | Type | Required | Notes |
|---|---|---|---|
| project | yes | — | |
| sort | string | no | What to order the endpoints by. requests is the default, and errors is the share of requests that returned 5xx. By the measurement, not by the text it is printed as, and a route with no requests i |
| order | string | no | Which way. By default the worst first for a measurement and alphabetical for a name, which is what the page does on the first click. |
What changed in a project since an instant: ask it first when a session starts, with the instant you
| Parameter | Type | Required | Notes |
|---|---|---|---|
| project | yes | — | |
| since | string | yes | RFC 3339 instant: when you last looked. Required, and not in the future. What happened at this instant was already seen and is not in the answer. |
| until | string | no | RFC 3339 instant: the end of the window, after since and not in the future. Leave it out: the answer fixes it at the instant it was prepared and gives it back in until and in next. What happened |
| limit | integer | no | How many items to return. Twenty by default. |
| offset | integer | no | The place of the first item to return, up to the window's total. 0 by default; next has it for the page after this one. |
The findings of a project, as one flat list: every open one, the most recently confirmed first, then every
| Parameter | Type | Required | Notes |
|---|---|---|---|
| project | yes | — |
One finding in detail: what was measured, under observed; what was deployed close enough before it to
| Parameter | Type | Required | Notes |
|---|---|---|---|
| project | yes | — | |
| finding | yes | — |
The report of a finding: facts, hypotheses with their state and evidence, the overall confidence
| Parameter | Type | Required | Notes |
|---|---|---|---|
| project | yes | — | |
| finding | yes | — |
What changed between the degraded window and its reference: the differences ordered by how much
| Parameter | Type | Required | Notes |
|---|---|---|---|
| project | yes | — | |
| finding | yes | — |
Did what I changed work? It reads the finding's footprint from since, the instant of the intervention,
| Parameter | Type | Required | Notes |
|---|---|---|---|
| project | yes | — | |
| finding | yes | — | |
| since | string | yes | RFC 3339 instant of the intervention. Required. |
What a project looked like further back than the fine-grained data goes, hour by hour. to must be
| Parameter | Type | Required | Notes |
|---|---|---|---|
| project | yes | — | |
| from | string | yes | RFC 3339 instant. |
| to | string | yes | RFC 3339 instant. |
| environment | string | no | The environment to read, in both windows. Without it, every environment together. |
| baselineFrom | string | no | RFC 3339 instant: the start of the other window, the one to compare against. Goes with baselineTo, both or neither. |
| baselineTo | string | no | RFC 3339 instant: the end of the other window, the one to compare against. Goes with baselineFrom, both or neither. |
Every error this project has observed, from the first one: its identity, how many times, when it
No parameter schema on file.
One error and where it was seen: every environment, deployed version and route it happened on, each
No parameter schema on file.
File an error as dealt with, attributed. That is all it does: it accepts no reference, silences no
No parameter schema on file.
Take an error out of the default list until a moment you choose, after which it comes back on its
No parameter schema on file.
Lift an ignore before its until, attributed: the error goes back to open and into the default list.
No parameter schema on file.
Say what you know about an error and Downtrace could not measure: 'only with the legacy checkout',
No parameter schema on file.
The newest hundred captures of a project, newest first, whatever their origin —requested, automatic or
No parameter schema on file.
One capture and whatever evidence has arrived, with both of its coverages, how the requests it kept spent
No parameter schema on file.
The requests one instance kept for a capture, a page at a time and in the order they arrived. Each comes
No parameter schema on file.
What this project says Downtrace missed: the newest hundred regressions recorded with record_regression,
No parameter schema on file.
Ask for detail on a route or a dependency for a while. Accepting is not observing: the answer says
No parameter schema on file.
Close a finding by hand, with one of the three reasons the product allows. This is not observed
No parameter schema on file.
Accept the current behaviour as the new normal for this finding. A finding that is already closed is
No parameter schema on file.
Record your own reading of a hypothesis, with its reason. It sits beside Downtrace's evaluation and never
No parameter schema on file.
Stop being told about something, with a scope and an end at most thirty days away. It silences the alert,
No parameter schema on file.
End a silence before its until, attributed to the credential that lifts it. Lifting one that was already
No parameter schema on file.
No parameter schema on file.
No parameter schema on file.
No parameter schema on file.
Free
$0 forever
- Every tool, no call limit
- No card, and mcprush charges nothing for it
- Source published under MIT
- 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 Downtrace 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 100 / 100
Release 0.3.1 · scanned 30 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 score100
Findings0 threat · 0 capability
Capabilityminimal
Coveragesource
Verificationsource
Surface digest94f624288ad97d2a26517cfa1c4b3571bff5269a3d702b8f2a986989989a054b
Score vectorthreat 100 → client 100
| Term | Level | What it prices | Points |
|---|---|---|---|
| capability-exposure | minimal | capability blast radius (minimal) — client exposure if the model is manipulated | −0 |
| verification-discount | source | publisher verification (provenance) — cryptographic build provenance ties the artifact to its source | −0 |
| 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 (29 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 |
|---|---|---|
| read_credential | no tags | |
| list_projects | no tags | |
| project_status | no tags | |
| changes_since | no tags | |
| list_findings | no tags | |
| read_finding | no tags | |
| read_report | no tags | |
| compare_windows | no tags | |
| verify_recovery | no tags | |
| read_history | no tags | |
| list_errors | no tags | |
| read_error | no tags | |
| resolve_error | no tags | |
| ignore_error | no tags | |
| unignore_error | no tags | |
| annotate_error | no tags | |
| list_captures | no tags | |
| read_capture | no tags | |
| read_captured_requests | no tags | |
| list_regressions | no tags | |
| request_capture | no tags | |
| close_finding | no tags | |
| accept_reference | no tags | |
| assess_hypothesis | no tags | |
| give_feedback | no tags | |
| annotate_finding | no tags | |
| record_regression | no tags | |
| silence_alerts | no tags | |
| lift_silence | 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 digest94f624288ad97d2a
Scanned2026-10-30
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 0.3.1. Pin a release below and the command asks for exactly that one.
0.3.1
undated
latestNo release note was published with this version.
—
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 Downtrace 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.