io.github.lmwharton/sieve-mcp
Sieve AI Due Diligence
AI startup due diligence — screen companies across 7 dimensions, get a Sieve Score.
alive probed 2026-10-15
4tools
753tokens in Claude Opus 5.5
801tokens in Claude Code
0.4 %of a 200k context window
396o200k_base estimate
9.5 sto initialize
665 msto list tools
- Run it
- uvx sieve-mcp==0.1.4 (stdio, probed in a container with the network off)
- Host
- pypi.org
- Reports itself as
- Sieve AI Due Diligence 4.1.0
- Capabilities
- logging, prompts, resources, tools
- Instructions
- 1 639 characters
- Claude count
- Claude Opus 5.5, measured 2026-10-15. Counted again when the tool definitions change. Method.
- Repository
- https://github.com/lmwharton/sieve-mcp
- Registry
- active, updated 2026-03-03 · registry entry
- Reachable
- 1 of the last 1 probe
Most expensive tools
| Tool | Tokens (o200k) | Share of this server |
|---|---|---|
| sieve_screen | 173 | 44 % |
| sieve_summary | 94 | 24 % |
| sieve_status | 80 | 20 % |
| sieve_usage | 49 | 12 % |
Badge and feed
[](https://mcpcost.ulehla.dev/s/io.github.lmwharton/sieve-mcp)
An entry for every probe that changed the result, the tool count or the token cost: Atom feed. Add it to a feed reader or a Slack RSS app to hear when the server stops answering or its tool set grows.