Skip to main content
Glama
yqi96

Warranted

by yqi96
OverviewSchemaRelated ServersScoreDiscussions
TypeScript
Local

Warranted

Make AI research agents accountable — give every conclusion a traceable argument graph.

Release License: MIT MCP Server Claude Code Plugin Stars

English | 简体中文

Coding agents have compilers and tests. Research agents have Warranted.

The problem

AI coding agents converge because compilers and tests provide objective failure signals. Research has no equivalent — conclusions live in natural language with no external verifier, so agents routinely declare work complete with no way to know what's missing.

Warranted fills that gap. It gives AI agents a persistent argument graph where every Claim requires Grounds and a Warrant, contradictions are recorded as Rebuttals rather than erased, and status can only advance after passing a logic check (compile). The result is research reasoning that is auditable, reproducible, and verifiable — not just plausible-sounding.

Good fit for: paper reproduction, hypothesis verification, multi-step scientific reasoning, research transparency.

Warranted Argument Map

Warranted — multi-tree overview

Related MCP server: Mcp-Omega-Brain

Documentation

New here? Read in this order:

Doc

What it covers

The Argument Graph

Core concepts — three node types and roles, compile, the status lifecycle, and how to talk to the agent in graph terms. Start here.

Reproducing a Paper

Scenario guide: verify a paper's claims with an independent argument graph (/paper-reproduce).

Writing a Paper

Scenario guide: draft a paper or literature survey where every citation traces to a verified Ground (/overleaf-setup, /literature-survey).

Release history: CHANGELOG.md

Setup with Claude Code

1. Clone and configure

Install Bun (>= 1.0.0), then:

git clone https://github.com/yqi96/warranted
cd warranted

# Optional: enable LLM logic review
cp review.json.example review.json
# Edit review.json and fill in apiKey

2. Register with marketplace (once per machine)

claude plugin marketplace add $(pwd)

3. Install the plugin in your project

cd your-project
claude plugin install warranted@warranted --scope local

On launch, toulmin-researcher becomes the primary agent and the MCP server starts automatically.

When LLM review is enabled, node definitions are reviewed on creation and compile_arguments runs a full logic-chain audit.

Hitting install or version issues? See known-working versions for a verified dependency snapshot.

Visualizer

From the warranted directory:

bun run viz

Open http://localhost:3456 in your browser.

Interaction

Action

Effect

Click a node

Select it (cyan glow ring)

Double-click a node

Open detail panel

Shift + click

Add to / remove from selection

Drag (box mode)

Draw a box to select multiple nodes

Drag (pan mode)

Pan the canvas

Scroll

Zoom

Click empty space

Clear selection

The ⬚ / ✥ buttons in the toolbar switch between box-select and pan mode.

Selection as context

The visualizer server tracks the current selection. Once the plugin is running, every message you send to Claude automatically includes the selected nodes as context — no need to describe which nodes you mean, just select and ask.

Agents

Agent

Role

toulmin-researcher

Primary agent. Builds and validates the argument graph, identifies structural gaps, drives each Claim toward a well-evidenced conclusion.

toulmin-explorer

Read-only. Quickly finds nodes, checks verification status, explores argument structure without making changes.

code-experimenter

Object-layer. Executes bounded coding, reproduction, and experiment tasks and returns evidence reports; does not decide Claim status.

discrepancy-auditor

Object-layer. Audits a negative outcome before it enters the graph — an unexpected mismatch about to become a Rebuttal, or a claimed blocker about to halt an obligation.

code-optimizer

Object-layer. Identifies hot paths, benchmarks, and implements targeted performance improvements within a bounded scope; does not decide Claim status.

Skills

Skill

Trigger

Role

paper-reproduce

/paper-reproduce

Paper reproduction workflow. Builds an independent argument graph and verifies paper claims step by step.

literature-survey

/literature-survey

Literature survey workflow. Grounds external findings in the argument graph, writes the survey in LaTeX with \cite{statement_N} citations. Maintains a .bib file throughout.

cite-review

/cite-review

Citation-faithfulness audit. Checks every \cite{statement_N} against its Statement in parallel, corrects mismatched LaTeX, and reconciles the graph.

academic-writing

/academic-writing

Manuscript-writing umbrella. Projects graph-backed arguments into paper prose across all sections, figures, tables, and citations.

overleaf-setup

/overleaf-setup

One-time setup skill. Installs leaf, authenticates, links a local LaTeX directory to an Overleaf project, and writes a Stop hook that auto-pushes on every conversation turn (skips if no files changed).

This server cannot be installed

A
license - permissive license
-
quality - not tested
B
maintenance

How are these scores calculated?

Maintenance

–Maintainers
–Response time
–Release cycle
1Releases (12mo)

Resources

Unclaimed servers have limited discoverability.

Looking for Admin?

If you are the server author, to access and configure the admin panel.

Related MCP Servers

View all related MCP servers

Related MCP Connectors

  • Penfield Memory

    Persistent memory and knowledge graphs for AI agents. Hybrid search, context checkpoints, and more.

  • Citation Verification

    Real-time fact-check, citation verification, and source-freshness for AI agents.

  • Touchstone

    Bitcoin-anchored, tamper-evident audit log for AI agents — record, disclose and verify actions.

View all MCP Connectors

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/yqi96/warranted'

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