MCP.Market

Contract Ops MCP server

by DrBaher·io.github.DrBaher/contract-ops-mcp·v0.3.0

One MCP server for the contract-ops suite: all nine local-first CLIs as agent tools.

A88/100grade A
What users say
No reviews yet
Be the first
Safety scan
A88/100

full report

Adoption
New

0 stars71 downloads/wk

Reviews

Write one

Nobody has reviewed Contract Ops yet.

If you have run it, two minutes of your experience saves the next person an afternoon.

Contract Ops tools (50, 2 write)

write = sends, deletes, buys or posts

Read from the package source without running it. The installed server may list more.

  • audit_show

    Show the hash-chained audit log for a sign request. Read-only.

  • catalog

    Return a CLI's full machine-readable command/flag catalog (`<cli> --catalog json`). Use this to discover the long tail beyond the curated tools.

  • compare_versions

    Clause-aware drift detection between two contract versions. exitCode: 0 clean · 2 substantive drift · 3 cosmetic · 4 clauses moved · 1 I/O error.

  • contract_vault_accept

    Mark register field(s) as manually verified (single, or bulk via a file). Writes to the register.

  • contract_vault_at_risk

    Renewal exposure: missed / imminent auto-renewal notice deadlines and expirations. Read-only.

  • contract_vault_due

    Project upcoming renewal / notice / payment deadlines from the signed-contract register.

  • contract_vault_export

    Export the register as csv | md | json (for spreadsheets / reports). Returns the export; writes no file. Read-only.

  • contract_vault_ingest

    Register a contract into the vault from an extract_contract JSON file. Writes to the register.

  • contract_vault_obligation

    Track one obligation's lifecycle (status / owner / recurrence / reminder days). Writes to the register.

  • contract_vault_obligations

    Project upcoming date/obligation actions from the register. Read-only.

  • contract_vault_query

    Query the register of signed contracts (read-only): list | find | get | show | stats | history.

  • contract_vault_remind

    Obligations whose reminder window is open right now — a digest for agents/cron. Read-only.

  • contract_vault_review

    List register fields needing review (unidentified / LLM-derived / low-confidence). Read-only.

  • contract_vault_risk

    Renewal-exposure analysis: missed/imminent auto-renewal notice deadlines and expirations.

  • contract_vault_verify

    Integrity check of the register (source sha256 + git state). Read-only.

  • convert_to_pdf

    Convert a Word document to PDF (needs a PDF backend such as LibreOffice on the host).

  • draft_nda

    Draft a new NDA from a built-in template into `out` (markdown; optional .docx). Deterministic. If placeholders are missing the result lists them — fill and retry. Needs nda_setup first.

  • extract_contract

    Ingest any contract (.md/.txt/.html/.docx/.pdf) into structured JSON — parties, dates, term, governing law, a clause map, defined terms — each with a confidence + source. Deterministic, no network.

  • fill_template

    Fill placeholders in a markdown/.docx template with parameter values (deterministic; no LLM). Returns the filled document on stdout.

  • generate_redlines

    Generate a clause-ready redline draft from a review_nda JSON report. Writes redline markdown to `out`. Deterministic; no org policy needed.

  • lint_contract

    Lint one contract for internal-consistency defects — leftover placeholders, broken cross-references, undefined/unused defined terms, numbering gaps, party/date inconsistencies. Returns findings (rule, severity, line).

  • nda_setup

    One-time: generate the org NDA policy + default playbook in the workspace (needed before draft_nda / negotiate_*). Idempotent; writes config into the workspace.

  • negotiate_accept

    Accept the current negotiated text, SIGNING convergence on your side. Signs a binding acceptance.

  • negotiate_analyzewrite action

    Post-hoc negotiation dashboard: trajectory, per-clause winners, source breakdown, outcome. Read-only.

  • negotiate_counter

    Sign a counter-round with amendments — either from a JSON amendments file you provide, or deterministic --auto with a stance. Signs a negotiating commitment.

  • negotiate_diff

    Clause-by-clause changes between two negotiation rounds (defaults to the last two). Read-only.

  • negotiate_finalize

    Finalize a converged negotiation: emit the final .md and .docx. The legal SIGNATURE stays with the human — no e-signature is applied here; hand off to sign-cli.

  • negotiate_init

    Start a negotiation: draft from a template + parties, SIGN as Party A, and write the state file. Signs a negotiating commitment. Needs nda_setup first.

  • negotiate_review

    Review the latest negotiation round against your policy. Read-only.

  • negotiate_status

    Show a negotiation's rounds, per-clause status, and signatures. Read-only.

  • negotiate_validate

    Integrity check on a negotiation state file: schema + hash-chain + per-round shape. Read-only.

  • review_nda

    Review an NDA against a house playbook — deterministic scoring with evidence per finding. Returns the structured review report (decision, risk_score, findings with evidence).

  • runwrite action

    Escape hatch: run any suite CLI with raw arguments (no shell). For commands the curated tools don't cover. Call `catalog` first to learn the flags. Note: signing stays human-gated — only sign-cli's read/verify subcommands are reachable here; request-create/send/sign/approve are rejected and must go through sign-cli's own MCP with its per-signer approval tokens.

  • suite_status

    Report which suite CLIs are installed (with versions) and how to install any that are missing. Call this first if a tool reports a CLI isn't installed.

  • template_vault_clause_library

    Find clauses that repeat across the vault (a reusable clause library). Read-only.

  • template_vault_clauses

    List the clauses detected in a template. Read-only.

  • template_vault_compare_clauses

    Compare clauses between two templates (optionally one named clause). Read-only.

  • template_vault_compose

    Fork a template into a new derived template in the vault. Writes a new versioned template.

  • template_vault_diff

    Unified diff between two versions of one template. Read-only.

  • template_vault_export

    Export a template to another format (e.g. .docx) at a workspace path.

  • template_vault_find

    Search the template vault by category, tag, jurisdiction, or keyword. Read-only.

  • template_vault_get

    Resolve and return a versioned template's text by reference (e.g. nda/house-mutual). Read-only.

  • template_vault_history

    Chronological timeline for a template: versions, swaps, and amendments. Read-only.

  • template_vault_info

    Show metadata for a template (category, tags, jurisdiction, versions). Read-only.

  • template_vault_list

    List templates in the vault, optionally filtered by category / tag / jurisdiction. Read-only.

  • template_vault_stats

    Vault dashboard: template counts, coverage, and last activity. Read-only.

  • template_vault_swap

    Replace one clause in a template with the same clause from another template. Writes a new version.

  • template_vault_verify

    Content-level sha256 integrity check of the vault. Read-only (never rewrites hashes).

  • verify_receipt

    Re-verify a portable signing receipt bundle, fully offline. Read-only.

  • verify_signature

    Verify a signed PDF matches what was recorded for its request. (Read-only; signing itself stays human-gated behind sign-cli's own MCP.)

Public scan report

scanner v0.1.9 · 2026-09-25 · same rubric, same numbers if you re-run it

no findings
  • Code scan3 source files scanned25/25
  • –Live reliabilityno gateway calls yet and no remote to proben/a
  • –Tool poisoningtools not inspected (local package is not executed); not countedn/a
  • Auth qualitylocal package, no credentials required12/15
  • Maintenancelast push 72 days ago12/15
  • Maintainer identityregistry namespace matches repository owner; GitHub account older than a year8/10
Overall 88/100. Components that don't apply are left out of the denominator. Any critical finding is an F.RubricAppeal a findingJSON

How Contract Ops compares

Against every graded AI and agents MCP server, from our own scans and uptime checks. Updated after each scan run.

Safety score
88/100
higher than 80% of 4,460 AI and agents servers
Tools
50
the typical AI and agents server has 6
Tools that change things
4%
typical: 0%. Fewer write tools means less an agent can break
Downloads a week
71
more than 55% of 1,538 AI and agents servers with a package

What its tools let an agent do: 47 read · 1 create · 1 change · 1 run or deploy

Install Contract Ops in Claude Code, Cursor or VS Code

Runs npx -y contract-ops-mcp on your machine. Read the scan report first; the gateway never runs local packages.

claude mcp add contract-ops-mcp -- npx -y contract-ops-mcp
Add to Cursor

What the publisher says

From the Contract Ops repository's README, as published. We do not edit it. Read it on GitHub

contract-ops-mcp

One MCP server for the whole contract-ops CLI suite — wire it up once and an agent (Claude, Cursor, Codex, …) gets all nine local-first CLIs as tools: extract, draft, lint, compare, convert, review, and the template + signed-contract vaults. Signing stays human-gated.

Run this

// e.g. Claude Desktop / Cursor MCP config
{
  "mcpServers": {
    "contract-ops": { "command": "npx", "args": ["-y", "contract-ops-mcp"] }
  }
}

The CLIs themselves must be installed (the server shells out to them). Fastest way to get all nine:

curl -fsSL https://cli.drbaher.com/install.sh | sh     # local
# …or run the server + CLIs in one container: ghcr.io/drbaher/contract-ops

Call suitestatus** any time to see which CLIs are present and how to install any that aren't.

Prefer a ready-made agent? contract-ops-agent is a

terminal agent that already wraps this server in an enclosure — the model's only tools are the contract-ops

tools, no shell/filesystem/signing by construction. Bring your own model (Claude, OpenAI, or any

OpenAI-compatible endpoint): npm i -g contract-ops-agent.

Tools

Curated, ergonomic tools (typed inputs, JSON out) for the common operations, plus two escape hatches for the long tail:

The curated set and catalog/run are discovery-driven — they ride the suite's uniform --catalog json contract, so they stay in sync as the CLIs evolve.

Safety

  • Signing is human-gated. Only sign-cli's read/verify operations are exposed (verifysignature, verifyreceipt, audit_show) — never request-create or sign. This is enforced: the run/catalog escape hatches reject any sign subcommand outside a read-only allowlist, so request-create/send/sign/approve can't be reached here. Those stay behind sign-cli's own MCP server with its per-signer approval tokens, so this server can't become an unguarded signing path.
  • Filesystem lockdown. File-path arguments to the curated tools are confined to CONTRACTOPSMCPBASEDIR (default: the working directory). Set it to widen the sandbox.
  • No shell. CLIs are invoked with execFile (no shell interpolation).

License

MIT. Part of the contract-ops CLI suite. See AGENTS.md for the agent contract.

Nothing above is checked by us. What we check is on the safety report.

Contract Ops: common questions

Is Contract Ops MCP server safe?
Yes, by our scan: it is graded A (88/100). Read the Contract Ops safety report
How do I install Contract Ops?
It runs on your machine. Copy the Claude Code, Cursor, VS Code or Claude Desktop config from the install section.
Does Contract Ops need an API key?
Not as far as the registry entry and our scan can tell: no credentials are declared or required.
Is Contract Ops maintained?
The last commit was 111 days ago (2026-07-16). The latest release is v0.3.0.
What can I use instead of Contract Ops?
Servers from other publishers that do the same job: Basic Memory MCP server, HOL Guard MCP server and Agent-Native Chat MCP server.
How does Contract Ops compare with other AI and agents MCP servers?
safety score 88/100 (higher than 80% of 4,460 AI and agents servers); tools 50 (the typical AI and agents server has 6); tools that change things 4% (typical: 0%. Fewer write tools means less an agent can break). Full comparison.

Alternatives to Contract Ops

Same job from other publishers: the closest match first, then the best rated.

  • Basic Memory
    Local-first knowledge management with bi-directional LLM sync via Markdown files.
    B
  • HOL Guard
    Local-first AI agent security evidence and approval workflows through HOL Guard's stdio MCP server.
    A
  • Agent-Native Chat
    Minimal chat-first app with durable threads, actions, and the app-agent loop
    A
  • DCC-MCP Core
    Agent control plane for DCC and engine hosts: one MCP endpoint, one tool contract.
    A
  • Agent Recall
    Correction-first agent memory. Precision KPI tracks if agents heed warnings. 5 layers, local-only.
    A

More from DrBaher →