Prooftask MCP Server
by cognitive-emergenceio.github.cognitive-emergence/prooftaskv0.1.2
Public template discovery over hosted MCP; scoped delivery workflows through the local adapter.
context tax
~332 tok/turn3 tools
security
pending
cold start
pending
freshness
pending
Our reading of Prooftask, from the data
Prooftask is an MCP server from cognitive-emergence, listed in the official MCP Registry.
Its 3 tools add about 332 tokens to every model turn (measured 2026-10-19).
It runs either locally over stdio or as a hosted endpoint.
Install Prooftask MCP server
Install in Claude Code
claude mcp add prooftask -- npx -y prooftask-mcpInstall in Cursor
{
"mcpServers": {
"prooftask": {
"command": "npx",
"args": [
"-y",
"prooftask-mcp"
]
}
}
}Add to ~/.cursor/mcp.json (global) or .cursor/mcp.json (project).
Install in Claude Desktop
{
"mcpServers": {
"prooftask": {
"command": "npx",
"args": [
"-y",
"prooftask-mcp"
]
}
}
}Settings → Developer → Edit Config (claude_desktop_config.json), then restart.
Install in VS Code
{
"servers": {
"prooftask": {
"type": "stdio",
"command": "npx",
"args": [
"-y",
"prooftask-mcp"
]
}
}
}Add to .vscode/mcp.json in your workspace.
Install in Windsurf
{
"mcpServers": {
"prooftask": {
"command": "npx",
"args": [
"-y",
"prooftask-mcp"
]
}
}
}Add to ~/.codeium/windsurf/mcp_config.json.
Remote endpoints
- streamable-http
https://prooftask-public-mcp-production.up.railway.app/mcp
Context tax
Connecting this server adds ~332 tokens of tool definitions to every model turn (lean, 3 tools). Measured 2026-10-19 via its hosted endpoint; o200k tokenizer, worst case for clients that load all tools up front. How we measure
| Model | Per 1,000 turns | With prompt caching |
|---|---|---|
| Claude Opus 5.5 | $1.33 | $0.066 |
| Claude Sonnet 5.5 | $0.664 | $0.066 |
| Claude Haiku 5.5 | $0.033 | — |
Input prices as of 2026-10-06. Excludes your own prompt and output.
Tools (3)
getPublicSpecification158 tokRead the exact public template version returned by searchPublicSpecifications. No credential needed. Review scope, completion criteria and required evidence with the user before recommending adoption. This read creates no draft, agreement or acceptance and never grants authority.
searchPublicSpecifications132 tokFind public delivery acceptance templates before asking a user to connect. Search by delivery scenario such as automation, customer support or software. No credential needed. Empty results mean broaden the query; do not invent a matching template. Returned examples are editable starting points, not quality certification or payment proof. Pin the returned version and digest before reuse.
getAgentCapabilities40 tokRead supported connection, billing and client capabilities without a credential
Freshness
Not enough public signals yet to judge maintenance.
Last commit (default branch)
2026-10-17 · 1d ago · GitHub
Latest release
no data · GitHub
Package published
no data · npm/PyPI
Registry entry updated
2026-10-17 · 1d ago · official registry · v0.1.2
FAQ
›How do I install the Prooftask MCP server in Claude Code?
Run: claude mcp add prooftask -- npx -y prooftask-mcp. For Cursor, VS Code, Claude Desktop and Windsurf, use the install tabs above.
›Does Prooftask require an API key?
No required environment variables are declared in its published metadata.
›Can I use Prooftask as a remote (hosted) MCP server?
Yes — it offers both a hosted endpoint and a local stdio package.
›Is Prooftask in the official MCP registry?
Yes, as io.github.cognitive-emergence/prooftask.
›How many tokens does Prooftask add to my context?
Its tool definitions total about 332 tokens, resent on every turn.
Sources and method
- Official MCP Registry entry
- updated 2026-10-17
- GitHub repository
- fetched 2026-10-19
- Package registry (npm / PyPI)
- listed
- Tool list (tools/list), measured by us
- 2026-10-19
- Last checked against all sources
- 2026-10-19
The description, packages and links come from the maintainer's own published metadata. Token cost, security signals, freshness and the reading above are MCPSkillHub's own measurements; see the methodology. Spot an error? Request a correction or claim the listing.