Skip to main content
Glama

Run Collection

run_collection

Execute requests in a Bruno collection and run test scripts, with support for subsets, isolated groups, parallel execution, and capturing variables from responses.

Instructions

Execute requests in a Bruno collection and run test scripts. HOW TO RUN A SUBSET: requests takes an ORDERED list of entries, each a .yml/.bru request file or a directory (which expands to every request under it, recursively). Absolute, or relative to collectionPath. Order is yours and duplicates are allowed — naming the same request twice runs it twice. Omit requests to run the whole collection. GROUPS: pass groups instead when one call needs more than one identity or configuration. Each group owns its OWN variable store and cookie jar, so nothing a group sets — a bru.setVar, a session cookie — is visible to any other group. That is what makes running the same five requests as alice and as bob in one call safe. Passing both requests and groups is REJECTED rather than resolved for you. PARALLELISM: parallel on the RUN fans the groups out against each other; parallel on a GROUP fans that group's own requests out. A group is serial unless it says otherwise, whatever the run does. seq no longer constrains execution — it is the default order and the reporting order only, so two requests in one parallel group genuinely run at the same time and may contend on the store they share. RESULTS are group-shaped: groups[] each carry their own summary, results, capturedVariableNames, capturedVariables and warnings, and the top-level summary is the run. There is no top-level results array, in the no-groups case either. A group that crashed outright reports error and counts as one failure in the run summary. Each result includes the response body (response_body, response_content_type, response_body_truncated) by default — disable with includeResponseBody=false or cap the size with maxResponseBodyBytes. Within a group, requests run in the order given; a directory expands by seq, scoped to its own folder, with subfolders before that directory's own loose requests, ties broken by filename. A request file that cannot be parsed is skipped when it was DISCOVERED — by running the whole collection, or by expanding a directory: the count is parseErrors and each skipped file is named with its reason in parseFailures. A file you NAMED yourself is different: you asked for that specific request and there is no partial answer to it, so it fails the group that named it, reported as that group's error, and the other groups still run. A named request that does not exist is reported in that group's missingRequests rather than failing anything, so you can see which subset ran. TRANSPORTS: http, graphql, grpc and websocket requests all run; any other kind is refused per-request with status 0 and a named reason, leaving the rest of the group alone. A gRPC result carries a grpc detail — the gRPC status code (0 is OK, and is NOT the refusal sentinel: status 0 with an error and no grpc detail is a refusal), the method, and redacted metadata. A WebSocket result carries a websocket detail — the transcript, stop_reason naming what ended the session (count, timeout, bytes, closed or error), and truncated, which is true for the three that cut a session short. Frame contents are recorded only if you ask for them, via websocket.includePayloads. Outbound requests are SSRF-filtered: targets resolving to private, loopback, link-local or otherwise reserved addresses are refused unless the server operator has allowlisted them, and a refusal is reported per-request as an "SSRF blocked" error with status 0.

Input Schema

TableJSON Schema
NameRequiredDescriptionDefault
groupsNoRun the same collection under more than one identity or configuration in one call. Each group owns its OWN variable store and cookie jar: nothing a group sets is visible to any other group, in either direction, at any parallel setting. Groups are reported in the order given. The same request may appear in several groups. Cannot be combined with requests; omit both to run the whole collection as one group.
parallelNoFan out. At the run level this runs the GROUPS concurrently; a group's own requests are serial unless that group sets its own parallel. With no groups given the run is one group, so parallel here runs every selected request concurrently. Reporting order is the listed order regardless. Default: false.
requestsNoThe requests to run, IN THE ORDER GIVEN. Each entry is a .yml or .bru request file, or a directory, which expands to every request under it, recursively. Absolute, or relative to collectionPath. Get paths from list_requests or get_collection_stats. Duplicates are allowed: naming the same request twice runs it twice. Omit to run every request in the collection; an empty [] is a selection of nothing and runs nothing. Cannot be combined with groups. An entry naming nothing is reported in missingRequests rather than failing the run, so you can see which subset ran.
cookieJarNoKeep cookies from each response and send them on later requests in the same run, so a login carries into the requests after it. Scoped to the group — nothing crosses from one group to another, at any parallel setting — held in memory, never written to disk, and matched by host/path/expiry — a cookie set by one host is not sent to another. Precedence, per cookie name: a Cookie header the request writes itself WINS over the jar, and the jar only adds names the request did not set, so a request that pins a specific credential keeps it and a run that relies on the jar is unaffected. This diverges from Bruno's CLI, where the stored value wins the clash; a warning names any cookie whose stored value was dropped. Default: true, matching Bruno's CLI. Set false to send only the Cookie headers a request writes itself.
variablesNoVariables for this run only, as {name: value}. They override the environment file and work without one. Held in memory and never written to any file — this is the only correct way to supply a secret, because neither Bruno file format stores a secret value. Referenced as {{name}} in urls, headers, bodies and auth. A request-level vars:pre-request entry or a bru.setVar in a script still overrides these, matching Bruno's --env-var precedence.
websocketNoBounds for websocket requests in this run. Applies to every websocket request in the call; there is no per-request form. Omit and the defaults below apply. A session always ends on one of these bounds or on the peer closing, and nothing is held open past the call.
environmentNoEnvironment name to use (e.g. "dev", "staging"). Get available names from get_collection_stats.
collectionPathYesAbsolute path to collection root directory. Use the path returned by list_collections.
collectionRootNoThe collection that collectionPath belongs to, when running a subfolder of one: environments and the collection- and folder-level scripts are resolved from here. Must be collectionPath itself or an ancestor of it — a root that does not contain the collection is rejected, because its root scripts would then run against these requests.
maxConcurrencyNoCeiling on requests in flight across the whole run. Omit to derive one from this machine's cores and memory, held below capacity. 0 lifts it entirely, at your own risk. Applies to THIS run and is given back when it ends, so it neither re-caps another run already in flight nor outlives the call that asked for it. A ceiling below the number of requests you meant to run at once silently serialises them, so reproducing a race needs a value at least as large as the number of racers.
captureVariablesNoNames of variables set by bru.setVar during the run whose values you want back, e.g. ["token"]. A script that captures a value out of a response — bru.setVar("token", res.body.token) — makes it available to later requests as {{token}}, and this is the only way to see it yourself; without it the value exists only inside the run. Every name a script set is listed in capturedVariableNames on every run, so run once to see what is there and then ask for the one you need. Values are only returned for names you list here, and they are returned verbatim. Nothing is captured from the environment file or from variables you supplied — only what a script set.
includeResponseBodyNoInclude the response body of each request in the results. Default: true.
maxResponseBodyBytesNoMaximum response body size (bytes) to return per request; longer bodies are truncated and response_body_truncated is set. Default: 10240.
Behavior5/5

Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?

With no annotations provided, the description carries the full burden and delivers exceptionally. It discloses SSRF filtering, parse-failure handling, group isolation of variable stores and cookie jars, parallelism semantics, cookie jar precedence, websocket bounds and security implications, and the exact shape of results. This goes well beyond what annotations could provide.

Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.

Conciseness4/5

Is the description appropriately sized, front-loaded, and free of redundancy?

The description is long but well-organized into uppercase sections (SUBSET, GROUPS, PARALLELISM, RESULTS, TRANSPORTS) and front-loaded with a clear purpose. Every sentence provides value, but the volume is high and may reduce quick readability. It is appropriately detailed for the tool's complexity, though not concise.

Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.

Completeness5/5

Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?

Given the tool's high complexity (13 parameters, nested objects, no output schema), the description is remarkably complete. It explains result shapes, error modes, concurrency, security filtering, and edge cases for every major path, ensuring an agent can invoke it correctly without additional assumptions.

Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.

Parameters5/5

Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?

Schema coverage is 100%, but the description adds substantial meaning beyond schema descriptions: directory expansion order, duplicate request handling, parseErrors vs named request failures, group-shaped results, transport-specific errors (gRPC, WebSocket), and SSRF refusal details. It enriches the understanding of many parameters.

Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.

Purpose5/5

Does the description clearly state what the tool does and how it differs from similar tools?

The description opens with 'Execute requests in a Bruno collection and run test scripts,' a specific verb+resource statement that clearly defines the tool's function. This distinguishes it from sibling tools like create_request, list_requests, or get_collection_stats, which handle other lifecycle operations.

Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.

Usage Guidelines4/5

Does the description explain when to use this tool, when not to, or what alternatives exist?

Provides explicit usage direction: when to use groups instead of requests ('GROUPS: pass groups instead when one call needs more than one identity or configuration'), how to run a subset, and that passing both requests and groups is rejected. It does not explicitly name alternative tools, but the internal usage guidance is strong.

Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.

Install Server

Other Tools

Latest Blog Posts

MCP directory API

We provide all the information about MCP servers via our MCP API.

curl -X GET 'https://glama.ai/api/mcp/v1/servers/Ostico/bruno-mcp-studio'

If you have feedback or need assistance with the MCP directory API, please join our Discord server