Skip to main content
Glama
vschernoff

forge-test-mcp

by vschernoff
OverviewSchemaSourceRelated ServersDiscussionsAdmin
JavaScript
Local

Server Configuration

Describes the environment variables required to run the server.

NameRequiredDescriptionDefault
rootNoProject directory (default: where the client starts the server). Passed as `--root <dir>`.
forgeNoThe `forge` executable (default: `forge` on `PATH`). Passed as `--forge <path>`.
timeoutNoStop a forge run after this many seconds (default 300). Passed as `--timeout <seconds>`.300
allow-ffiNoLet tests use FFI cheatcodes when `foundry.toml` enables them. Off by default. Passed as `--allow-ffi`.false
strict-versionNoRefuse forge versions newer than the tested range instead of flagging them. Passed as `--strict-version`.false

Instructions

Guidance the server publishes about itself, which clients place ahead of the tool catalog so the model reads it before choosing anything.

This server publishes no instructions, or was last inspected before Glama recorded them.

Capabilities

Features and capabilities supported by this server

Protocol revision2025-11-25

CapabilityDetails
tools
{
  "listChanged": true
}

Tools

Functions exposed to the LLM to take actions

NameDescription
run_testsA

Run forge test and return a compact summary: counts, and for each failing test its kind (unit, fuzz, invariant, setup), decoded reason, fuzz counterexample or invariant call sequence, and console logs. isError means the tests did not run (compile error, no matching tests, timeout, missing forge, …); never report that as passing.

explain_failureA

Re-run one test at high verbosity and return only that test: its reason, decoded call trace (trimmed to the path to the revert when long), backtrace, logs and, for fuzz or invariant tests, the counterexample. Pass the test name as run_tests reports it, e.g. "BankTest.test_transfer()".

buildA

Compile the project with forge build and return compiler errors and warnings as structured items (file, line, column, code, message). Unchanged files are not recompiled and forge does not repeat their warnings; pass force: true to recompile everything.

Prompts

Interactive templates invoked by user choice

NameDescription

No prompts

Resources

Contextual data attached and managed by the client

NameDescription

No resources

Deploy Server
Shareable, User-Owned Context BackendPeople, applications, and agents read, write, and coordinate together in one context lake. Reach it over MCP, a CLI, a Python SDK, or REST.fulcradynamics.comAd

TDQS

A4.1/5.0

Scored across 3 tools

Disambiguation5/5

run_tests, explain_failure, and build have clearly distinct purposes: one runs the suite, one drills into a single failure at high verbosity, one compiles. An agent can easily route each request.

Naming Consistency4/5

run_tests and explain_failure follow verb_noun, while build is a bare verb. Minor deviation but still readable and predictable.

Tool Count4/5

Three tools is slightly thin but each earns its place covering the core workflow of build, run, and diagnose. Well-scoped for a focused Foundry testing server.

Completeness4/5

The build-run-explain lifecycle is fully covered, including structured compiler output and fuzz/invariant counterexamples. Minor gaps like listing tests or coverage exist but core workflows are complete.

Maintenance

ActivityMaintained
ResponsivenessNo issues

Resources