Ticket MCP Server
Enables AI agents to manage a file-backed ticketing system directly within a local repository using a structured state machine and directory hierarchy. It enforces strict markdown schemas and provides specialized tools for claiming tasks, appending work logs, and validating ticket metadata.
README
Ticket MCP Server
Local MCP server for the repo ticket system (file-backed).
Features
- List/get/update/move/validate tickets in
tickets/ - Create new tickets with strict frontmatter validation
- Agent helpers:
tickets_next_id,tickets_claim,tickets_append_worklog,tickets_reconcile - Strict schema enforcement (rejects invalid writes)
- HTTP MCP endpoint plus optional stdio proxy
Directory Structure & State Machine
The MCP server expects a tickets/ directory in the root of your project (configured via TICKET_ROOT), containing the following subfolders that directly map to the ticket's status field. The server will strictly enforce that tickets reside in the correct folder for their status:
tickets/pending/→status: pending(Ready to be picked up)tickets/in_progress/→status: in_progressorstatus: blocked(Claimed and being worked)tickets/awaiting_human_test/→status: awaiting_human_test(AI work complete; needs human testing)tickets/done/→status: done(Human has verified and accepted)tickets/archive/→status: archived(Obsolete or old tickets)
Strict Markdown Validation
To prevent AIs from deleting context or inventing their own ticket formats, the MCP server enforces that all tickets contain specific markdown headers in their body. These headers will be automatically injected when creating a ticket via tickets_create.
Required body headers:
## Overview## Approach (medium/high-level)## Tasks / Todos## Requirements (AI implementation)## Human Testing Steps## Key Files / Areas (notes)## Questions## Blockers## Implementation Notes
Usage
Build:
npm install
npm run build
Start HTTP server:
npm run start
Auto-rebuild on changes (dev watch):
npm run dev
Run integration tests:
npm run test
Start stdio proxy (for MCP clients expecting stdio):
npm run stdio
Auto-restart stdio proxy on changes:
npm run stdio:dev
Note: build output lands under dist/ticket-mcp/src/ because the build
includes shared schema sources.
Configuration
TICKET_ROOT(default: repo root inferred from cwd)TICKET_MCP_PORT(default: 3334)TICKET_MCP_PATH(default: /mcp)TICKET_STRICT(default: true)
tickets_stats response
The tickets_stats tool returns aggregate counts plus ticket numbering metadata:
{
"status": {
"pending": 12,
"in_progress": 3,
"done": 25
},
"area": {
"tooling": 10,
"combat": 8
},
"epic": {
"none": 15,
"qa": 4
},
"highest_ticket_number": 42,
"next_ticket_number": 43
}
highest_ticket_number: highest numeric suffix parsed from existing ticket IDs (or0if none are numeric).next_ticket_number: convenience value equal tohighest_ticket_number + 1.
Agent helper tools
-
tickets_next_id- Returns
highest_ticket_number,next_ticket_number, andsuggested_id. - Optional args:
prefix(defaultT),separator(default-),padding(default3).
- Returns
-
tickets_claim- Claims a
pendingticket, moves it toin_progress, sets claim metadata, and appends awork_logentry. - Requires
actor; acceptsidorpath.
- Claims a
-
tickets_append_worklog- Appends a validated
work_logentry and updatesupdated_at. - Requires
entry; acceptsidorpath.
- Appends a validated
-
tickets_reconcile- Audits ticket invariants and optionally applies safe fixes.
- Optional
apply_fixes: truewill fix common metadata/date issues and folder/status mismatches when possible.
Recommended agent workflow
Use this sequence for normal implementation flow:
// 1) Allocate a canonical new id
const next = tickets_next_id({ prefix: "T", separator: "-", padding: 3 })
// 2) Create the ticket with suggested id
tickets_create({
id: next.suggested_id,
title: "Example ticket",
area: "tools",
epic: "none",
intent: "Why this work matters",
requirements: ["Requirement A"],
human_testing_steps: ["Do X"],
constraints: ["Constraint Y"],
key_files: ["tools/ticket-mcp/src/server.ts"],
status: "pending"
})
// 3) Claim when work starts
tickets_claim({
id: next.suggested_id,
actor: "worker-ai:example",
summary: "Starting implementation"
})
// 4) Add progress notes during execution
tickets_append_worklog({
id: next.suggested_id,
entry: {
actor: "worker-ai:example",
kind: "change",
summary: "Implemented the core logic"
}
})
// 5) Validate and reconcile before handoff
tickets_validate({ id: next.suggested_id })
tickets_reconcile({ id: next.suggested_id }) // preview only
Recommended Servers
playwright-mcp
A Model Context Protocol server that enables LLMs to interact with web pages through structured accessibility snapshots without requiring vision models or screenshots.
Magic Component Platform (MCP)
An AI-powered tool that generates modern UI components from natural language descriptions, integrating with popular IDEs to streamline UI development workflow.
Audiense Insights MCP Server
Enables interaction with Audiense Insights accounts via the Model Context Protocol, facilitating the extraction and analysis of marketing insights and audience data including demographics, behavior, and influencer engagement.
VeyraX MCP
Single MCP tool to connect all your favorite tools: Gmail, Calendar and 40 more.
Kagi MCP Server
An MCP server that integrates Kagi search capabilities with Claude AI, enabling Claude to perform real-time web searches when answering questions that require up-to-date information.
graphlit-mcp-server
The Model Context Protocol (MCP) Server enables integration between MCP clients and the Graphlit service. Ingest anything from Slack to Gmail to podcast feeds, in addition to web crawling, into a Graphlit project - and then retrieve relevant contents from the MCP client.
E2B
Using MCP to run code via e2b.
Neon Database
MCP server for interacting with Neon Management API and databases
Exa Search
A Model Context Protocol (MCP) server lets AI assistants like Claude use the Exa AI Search API for web searches. This setup allows AI models to get real-time web information in a safe and controlled way.
Qdrant Server
This repository is an example of how to create a MCP server for Qdrant, a vector search engine.