A command-line test harness for checking an MCP server, the protocol that lets tools communicate with an agent. It records tool requests and responses and produces a standard verdict.

Testing an MCP server end to end, checking its tool capabilities, and investigating reported MCP issues.

It helps find broken tool contracts and MCP connectivity problems with recorded evidence.

mcp-harness — Skill for Claude Code and Codex by sabbour

Install

Getting it into your agent

It runs from inside its repository, so the clone comes first — what it calls does not travel with the file alone.

Clone the repo

git clone --depth 1 https://github.com/sabbour/agentweaver

agentmods

npx agentmods add skills/sabbour/agentweaver/mcp-harness

Made for: Claude Code, Codex.

Wrote this? Show the measurements

A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.

README.md

[![agentmods](https://agentmods.dev/badge/skills/sabbour/agentweaver/mcp-harness.svg)](https://agentmods.dev/skills/sabbour/agentweaver/mcp-harness)

Your own site

<a href="https://agentmods.dev/skills/sabbour/agentweaver/mcp-harness"><img src="https://agentmods.dev/badge/skills/sabbour/agentweaver/mcp-harness.svg" alt="Measured on agentmods" height="20"></a>

Per session 0 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.

When invoked 2,265 The whole file, excluding the scripts and references it only reads on demand.

Security scan A 0 findings. A grade says what 26 rules found in the file — not that it is safe.

Origin original No closer match found in the catalogue.

Token cost

What it costs to keep this loaded

Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.

Model	Per session	Once invoked
Fable 5.1	$0.00000	$0.02265
Opus 5	$0.00000	$0.01132
Sonnet 5	$0.00000	$0.00453
Haiku 4.5	$0.00000	$0.00227

Measured 2d ago against content hash 4eba3cd0b132, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-07, from the pricing page.

Security

Grade A, and why

mcp-harness scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 2d ago.

The scan reads SKILL.md. This mod also ships 24 executable files (agent-driver/tools.mjs, convert-cli-native.mjs, lib/approval-safety.mjs, …), listed below but not scanned — reading those needs a real analyzer, not pattern matching.

A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.

Nothing flagged

None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.

scripts/mcp-harness/SKILL.md · 182 lines

How it starts

The opening of the file, as written. The whole thing — 182 lines — stays where its author put it; the contents beside it link to each section on GitHub.

MCP harness CLI contract

Use this harness to validate Agentweaver's MCP protocol surface, capture the complete tools/call request/response evidence, and emit a normalized agentweaver.persona-judge-verdict/v1 JSON verdict. It is for MCP end-to-end validation, MCP tool-contract regression checks, and investigation of an MCP-reported issue; use the API or UI harness for those surfaces.

Run all commands below from the repository root. The harness requires Node 18 or newer. Install its dependencies and run its unit tests before changing it:

npm --prefix scripts/mcp-harness install
npm --prefix scripts/mcp-harness test

There are two entry points, for two different jobs:

run-persona.mjs — full dynamic persona scenario validation (no fixed script; a dispatched sub-agent drives the live tool menu). Use this to validate how the MCP surface behaves for a real persona.
smoke/mcp-cli-smoke.mjs — a fast, fixed deterministic connectivity + capability tripwire. Use this for a quick "is the server reachable and does it still satisfy the capability contract" check.

Both transports work the same for either entry point:

stdio (--target stdio) spawns a local subprocess via --server-command/--server-args. There is no network target validation — stdio is only a transport selector, never a URL.
http (--target <url>) requires a real base URL that must include the exact /mcp pathname (e.g. https://<host>/mcp; the bare origin and /mcp/ are not the endpoint). Any HTTPS host is accepted; HTTP is loopback-only. URL credentials/fragments and TLS bypasses are rejected. http transport also requires OAuth: transient AGENTWEAVER_TOKEN must be an Agentweaver broker token for the exact /mcp resource with mcp:invoke, obtained through the app's OAuth flow. Raw Entra and GitHub tokens are rejected. Stdio transport still uses AGENTWEAVER_TOKEN for downstream API calls.

Driving a persona scenario (the only way — dynamic, no fixed scripts)

Read the full file on GitHub · 182 lines

Files

What ships with it

33 files beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.

Changes

What this file has done since we first saw it

Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.

2d ago Changed · +8 lines 4eba3cd0b132
4d ago First seen · 174 lines · 0 tokens per session scan A 896acf3ecc86

Subscribe to this mod's changes

mcp-harness is a skill published in the GitHub repository sabbour/agentweaver (5 stars, last pushed 2d ago), licensed MIT. It costs nothing until one of its globs matches a file; then it loads 2,265 tokens. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-09-03.

Debugging Testing agent-orchestration ai-agents aks dotnet human-in-the-loop kubernetes

routing-calibration-loop

Calibrates routing changes against a live router endpoint with executable probes, local DSL validation, versioned deploys, and structured failure review. Use when tuning signals, projections, decisions, or maintained route examples against a real apiserver.

vllm-project/semantic-router · 53 tokens

behavior-contract

Bug condition/postcondition formalization as testable Behavior Contracts. Defines invariants that must be preserved across fixes.

a5c-ai/babysitter · 25 tokens

quality-hooks

Language-specific auto-lint/format/typecheck pipeline. Supports Python (ruff+pyright), TypeScript (prettier+eslint+tsc), Go (gofmt+golangci-lint). Auto-fix and convergence loops.

a5c-ai/babysitter · 53 tokens

ward

Ward the codebase — an ad-hoc test-gap audit that reveals unguarded code (untested functions, uncovered edge cases, missing error-handling and integration tests) and conjures ready-to-paste test skeletons as protective wards against regression, surfaced inline in the session. Scoped to your current diff by default, or…

eric-cielo/moflo · 0 tokens

dos-self-improve

Run the DOS self-improvement loop: propose a candidate, verify it in an isolated worktree, measure suite/truth/metric gates, and keep only confirmed gains. Use for recursive improvement; use dos-enforce-tune for policy knobs.

anthony-chaudhary/dos-kernel · 56 tokens

test-runner

Use this skill when orchestrating agentic end-to-end tests. Resolves target + profile, dispatches the right driver(s) (playwright for web today, peekaboo for macOS (issue #381)), invokes the ux-evaluator agent (opus, read-only) against driver artifacts, reconciles findings with the open issue tracker via…

Kanevry/session-orchestrator · 124 tokens