Skip to main content
Glama
archimedes-market

Archimedes Market MCP

Official

@archimedes-market/mcp

Ask before you install.

Your coding agent picks dependencies on your behalf, often without you reading a line of them. This MCP server lets it check whether those packages are actually safe first — against Archimedes, which scores public GitHub repositories for security, code quality, license health and completeness (0–100 each) across a corpus of 3,000+ repos that grows daily.

> add a date library to this project

  check_dependencies(["date-fns", "moment", "dayjs"])

  ⚠ 1 dependency warrants review before shipping:
  - moment/moment: trust 61/100, security 48/100 — archimedes.market/r/moment/moment

Trust tools

  • check_dependencies(packages[]) — the main one. Pass npm package names ("express", "@scope/pkg") or GitHub coordinates ("owner/repo"), up to 30. Returns overall + security score, trust badge and a report link for each. Call it before installing packages, when reviewing a package.json or lockfile, or when choosing between libraries. npm names are resolved to their GitHub repo automatically.

  • check_repo_trust(owner, repo) — the full report for one repository: per-dimension scores, finding counts by severity, headline findings, detected licence, language, stars.

  • search_trust_reports(query?, asset_class?, min_score?) — search the scored corpus. "What are the most trustworthy MCP servers?" Best-scoring first.

An unscanned repo is not a bad repo. If Archimedes hasn't looked at something yet, the tool says not_scanned and gives you a URL to scan it free — it never implies absence of a report means absence of quality.

Related MCP server: taskbounty-mcp-server

Bounty tools

  • search_bounties — open engineering bounties by query, category, funding status, price band.

  • get_bounty_details(id) — full requirements, deliverables and acceptance tests for one bounty.

  • get_platform_stats() — aggregate counters. Cached upstream 60s.

Every bounty is funded in Stripe escrow before engineers see it, and every submission is AI-verified (Semgrep + LLM code review + licence scan) before the buyer sees it.

Install

Claude Desktop

Add to ~/Library/Application Support/Claude/claude_desktop_config.json (macOS) or %APPDATA%\Claude\claude_desktop_config.json (Windows):

{
  "mcpServers": {
    "archimedes": {
      "command": "npx",
      "args": ["-y", "@archimedes-market/mcp"]
    }
  }
}

Restart Claude Desktop. The three tools (search_bounties, get_bounty_details, get_platform_stats) will appear in the available tools list.

Cursor

Settings → MCP → Add server. Use the same config block as above.

Continue / other stdio clients

Point the client at:

npx -y @archimedes-market/mcp

Non-stdio clients (server-side agents, hosted bots)

Skip this package entirely — call the hosted HTTP endpoint directly:

POST https://archimedes.market/api/mcp
Content-Type: application/json

{
  "jsonrpc": "2.0",
  "id": 1,
  "method": "tools/call",
  "params": {
    "name": "search_bounties",
    "arguments": { "query": "MCP server", "limit": 5 }
  }
}

Full docs: https://archimedes.market/mcp

Example agent queries

"Find me open Archimedes bounties for KiCad PCB review under $3,000."

"Are there any MCP-server bounties on Archimedes right now? Show me the top 5 by payout, then give me the full requirements for the highest-paying one."

"How active is Archimedes right now? How much has been paid out?"

The tools return both a human-readable text block (for the model to reason over) and a structured payload (for downstream tooling).

Configuration

Env var

Default

Purpose

ARCHIMEDES_PUBLIC_API_URL

https://archimedes.market

Override the upstream base URL (preview deployments, local dev)

ARCHIMEDES_MCP_USER_AGENT

mcp-archimedes/0.2 (+https://archimedes.market)

Override the User-Agent sent on outbound calls

Health check

Verify the bridge can reach the upstream API before wiring it into a client:

npx @archimedes-market/mcp --probe

Exit code 0 means upstream is reachable. Non-zero with stderr diagnostic on failure.

Privacy

The Archimedes public API logs query_hash (SHA-256 of normalized params) and ip_hash (HMAC with daily-rotated salt) per call — never raw queries or raw IPs. 90-day retention. Zero-result queries are aggregated to inform what bounties Archimedes should source next.

License

MIT. See LICENSE.

Available Tools

1 tool
search_bountiesAInspect

Search open bounties on Archimedes Market — a marketplace for verified engineering work (software, hardware, MCP servers, CAD, EDA). Returns a paginated list with bounty title, summary, payout in cents, deadline, and a public URL. Use this to discover paid work that matches an agent's capabilities, or to surface engineering tasks a user can post a submission to.

ParametersJSON Schema
NameRequiredDescriptionDefault
queryNoFree-text search across bounty title and description. Use plain English — no SQL, no wildcards.
categoryNoMission type filter. Common values: "software", "hardware", "creative", "research", "mcp". Omit to include all categories.
statusNo"open" = biddable now (escrow locked). "funded" = any bounty that touched real money (locked OR released). "all" = no status filter, includes unfunded drafts.open
min_price_centsNoMinimum bounty payout in cents (USD).
max_price_centsNoMaximum bounty payout in cents (USD).
limitNoPage size, 1–50.
offsetNoPagination offset.

TDQS

A4.2/5.0
Behavior3/5

Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?

No annotations are provided, so the description carries the full burden. It mentions that results are paginated and lists return fields, but it does not disclose behavioral details such as default sorting order, rate limits, whether the search is fuzzy or exact, or how deadlines are formatted. More transparency would be helpful.

Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.

Conciseness5/5

Is the description appropriately sized, front-loaded, and free of redundancy?

The description is just two sentences, each earning its place. The first sentence defines the tool and its return fields; the second provides usage guidance. It is front-loaded and contains no filler.

Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.

Completeness4/5

Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?

Given 7 parameters, no output schema, and no sibling tools, the description covers the essential: what the tool does, what it returns, and when to use it. It could be improved by mentioning default ordering (e.g., 'by deadline' or 'by relevance') or whether the paginated list is sorted. However, it is largely complete for a search tool.

Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.

Parameters4/5

Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?

The input schema has 100% description coverage, so the baseline is 3. The description adds value by clarifying the 'query' parameter usage ('plain English — no SQL, no wildcards') and explaining the 'status' enum values ('open' = biddable now, 'funded' = any with real money, 'all' = unfunded drafts). This extra context justifies a score above baseline.

Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.

Purpose5/5

Does the description clearly state what the tool does and how it differs from similar tools?

The description clearly states the tool searches 'open bounties on Archimedes Market' and specifies the exact fields returned ('bounty title, summary, payout in cents, deadline, and a public URL'). The verb 'search' and the context of a marketplace for engineering work make the purpose unambiguous.

Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.

Usage Guidelines4/5

Does the description explain when to use this tool, when not to, or what alternatives exist?

The description explicitly states when to use this tool: 'to discover paid work that matches an agent's capabilities, or to surface engineering tasks a user can post a submission to.' While it does not mention when not to use it, the absence of sibling tools makes this less critical. It provides clear context for appropriate usage.

Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.

TDQS

A4/5.0
Disambiguation5/5

Only one tool exists, so no ambiguity between tools. The singular tool has a clear distinct purpose.

Naming Consistency5/5

With only one tool named 'search_bounties', the naming follows a predictable verb_noun pattern and is perfectly consistent.

Tool Count2/5

A marketplace server with only one search tool feels too few for the apparent scope. Agents likely need additional tools for details, creation, or submissions.

Completeness2/5

The surface is severely incomplete: only search is available, with no way to view full bounty details, create or update bounties, or submit work, causing agent failures.

Maintenance

ActivitySlowing
ResponsivenessNo issues

Resources

Unclaimed servers have limited discoverability.

Looking for Admin?

If you are the server author, to access and configure the admin panel.

Related MCP Connectors

Related MCP Servers

Latest Blog Posts

MCP directory API

We provide all the information about MCP servers via our MCP API.

curl -X GET 'https://glama.ai/api/mcp/v1/servers/archimedes-market/mcp-archimedes'

If you have feedback or need assistance with the MCP directory API, please join our Discord server