mcp-graph-workflow

An MCP-enabled, local-first CLI that ingests PRD documents and converts them into persistent, structured execution graphs stored in SQLite. It supports AI-assisted context compression, semantic search/RAG, task routing, sprint planning, and a local web dashboard; it also coordinates multi-agent workflows via MCP tool calls.

Evaluated Mar 30, 2026 (0d ago)
Repo ↗ DevTools mcp task-graph prd sqlite local-first rag semantic-search typescript cli orchestration
⚙ Agent Friendliness
59
/ 100
Can an agent use this?
🔒 Security
32
/ 100
Is it safe for agents?
⚡ Reliability
32
/ 100
Does it work consistently?

Score Breakdown

⚙ Agent Friendliness

MCP Quality
70
Documentation
65
Error Messages
0
Auth Simplicity
90
Rate Limits
10

🔒 Security

TLS Enforcement
30
Auth Strength
20
Scope Granularity
20
Dep. Hygiene
55
Secret Handling
45

MCP is used via local stdio with no auth described. The REST/dashboard presence is noted but authentication, TLS requirements for the HTTP server, CSRF/CORS controls, and secret-handling/logging behavior are not documented in the provided content. Dependencies include common libraries for parsing and browser automation; no vulnerability posture/CVE hygiene information was provided.

⚡ Reliability

Uptime/SLA
0
Version Stability
55
Breaking Changes
40
Error Recovery
35
AF Security Reliability

Best When

You want local, offline-friendly PRD-to-task-graph scaffolding and agent tool orchestration via MCP (stdio), plus a browser dashboard for planning/insights.

Avoid When

You require robust enterprise security controls around authentication/authorization, or you need well-specified rate-limit/error semantics for a REST API as the primary integration surface.

Use Cases

  • Convert PRD/spec text into dependency-aware, trackable task graphs for agentic execution
  • Local semantic search and RAG over PRDs/backlogs to ground agents
  • Sprint planning and capacity/risk assessment from imported requirements
  • Operate MCP-native agent workflows from editors (Copilot/Claude/Cursor/etc.) using stdio transport
  • Use a dashboard to visualize graphs, backlog progress, and insights

Not For

  • Centralized, multi-tenant SaaS use where strong auth, tenancy isolation, and audit trails are required
  • Workloads needing managed cloud storage, remote vector databases, or external knowledge bases
  • Teams that require a fully specified OpenAPI/SDK-driven REST integration for automation

Interface

REST API
Yes
GraphQL
No
gRPC
No
MCP Server
Yes
SDK
No
Webhooks
No

Authentication

Methods: No auth described for local MCP stdio transport Local REST/dashboard (auth behavior not documented)
OAuth: No Scopes: No

The README shows MCP stdio usage via npx with no credentials. It mentions REST endpoints and a dashboard but does not describe authentication/authorization mechanisms or access controls.

Pricing

Free tier: No
Requires CC: No

Open-source MIT project; pricing not applicable based on provided content.

Agent Metadata

Pagination
none
Idempotent
False
Retry Guidance
Not documented

Known Gotchas

  • MCP tooling is stdio-based; agent environments must support launching/stderr/stdin transport correctly
  • Local execution implies filesystem/permissions issues can occur; 'doctor' exists but agent retry/error handling guidance is not described
  • Graph/database state may persist in SQLite; repeated imports/updates might create duplicates unless tools are designed to be idempotent (not confirmed)

Alternatives

Full Evaluation Report

Comprehensive deep-dive: security analysis, reliability audit, agent experience review, cost modeling, competitive positioning, and improvement roadmap for mcp-graph-workflow.

AI-powered analysis · PDF + markdown · Delivered within 30 minutes

$99

Package Brief

Quick verdict, integration guide, cost projections, gotchas with workarounds, and alternatives comparison.

Delivered within 10 minutes

$3

Score Monitoring

Get alerted when this package's AF, security, or reliability scores change significantly. Stay ahead of regressions.

Continuous monitoring

$3/mo

Scores are editorial opinions as of 2026-03-30.

6533
Packages Evaluated
19870
Need Evaluation
586
Need Re-evaluation
Community Powered