mcpField
R
release-to-tasks

release-to-tasks

Local
github.com/apomtDeveloper Tools0Updated 9w ago

About release-to-tasks

release-to-tasks is an MCP server for turning public release-note URLs into ordered upgrade tasks. Each task includes a source excerpt, URL anchor, confidence, and uncertainty, while the server avoids executing release text or project code.

Key features

  1. 01Parse public release-note sources
  2. 02Produce ordered upgrade tasks
  3. 03Include source excerpts and anchors
  4. 04Mark confidence and uncertainty
  5. 05Reject unsafe URLs and redirects

Use cases

  1. 01Review upgrade work from GitHub releases
  2. 02Turn GitLab release notes into task lists
  3. 03Assess changelog items before applying upgrades

MCP support

The MCP server provides the `release_url_to_tasks` tool, which accepts a public release-note URL and returns source-cited tasks and warnings.

Configuration

Local server config is the same for every client. Copy it into your client’s MCP config file.

mcp.json · Universal config
{
"mcpServers": {
"apomt-release-to-tasks": {
"command": "npx",
"args": [
"-y",
"release-to-tasks-mcp"
]
}
}
}
Run commandnpx -y release-to-tasks-mcp

Local servers need a runtime on your machine (e.g. Node.js 18+). Config file locations differ by client; see your client’s docs.

Packages
npmrelease-to-tasks-mcp0.1.0

Tools

1

Saved tool list · updated 3w ago

ParameterTypeRequiredDescription
release_urlstringYesPublic HTTPS GitHub, GitLab, Markdown, or HTML release-note URL

Details

From the project README

release-to-tasks-mcp

Turn one public release-note URL into ordered upgrade tasks that retain a source excerpt, anchor, confidence, and uncertainty. The MCP server never executes release text or project code.

Install

Node.js 22.14 or newer is required.

{
  "mcpServers": {
    "release-to-tasks": {
      "command": "npx",
      "args": ["-y", "[email protected]"]
    }
  }
}

For a clean local invocation before npm publication:

npm ci
npm test
node dist/index.js

Tool

release_url_to_tasks

{
  "release_url": "https://github.com/modelcontextprotocol/python-sdk/releases/tag/v2.1.1"
}

The result contains sourceFormat, releaseTitle, ordered tasks, and warnings. Every task includes the exact source URL/anchor and a short source excerpt. Low-confidence entries explicitly require human impact assessment.

Supported sources:

  1. GitHub release pages (resolved through the public GitHub API)
  2. GitHub release API URLs
  3. GitLab release pages
  4. Public Markdown/plain-text changelogs
  5. Public HTML release-note pages

Malformed URLs, non-HTTPS URLs, local/private destinations, unsupported pages, missing release bodies, oversized responses, and unsafe redirects return explicit errors.

Demo

  1. Configure the server in an MCP client.
  2. Call release_url_to_tasks with a real public release URL.
  3. Check each proposed task against its source.excerpt and source.url before applying an upgrade.

The repository includes a non-fabricated sample response shape. The sample deliberately contains no tasks because release content can change; tests contain deterministic source fixtures.

Privacy and security

Telemetry is disabled by default and there is no hosted collector. Optional local aggregate telemetry never includes URLs or release contents. See PRIVACY.md and SECURITY.md.

Future extension points

The parser and fetcher boundaries can later support authenticated private-source connectors. This release contains no payment, Stripe, private-source, or hosted-service dependency.

Development

npm ci
npm test
npm pack --dry-run

Internal tests and owner traffic are technical signals only; they never count as external adoption. LEVEL 2 requires three unrelated external users to install the public package and successfully process real release URLs.

License

MIT

Comments

Markdown supported

New to MCP?

From what the protocol is to installing your first server, these answers cover most first questions.

An open protocol that defines how AI applications connect to external tools and data in a uniform way. Clients talk to the model; MCP servers provide tools, resources and prompt templates. They exchange standard messages, so one server can be reused by many clients.