# nanotoll (nanotoll.dev)

Compresses LLM agent context via deterministic pruning, extractive token selection, or NLI-verified squeezing, billed per call.

- Atlas card: https://inite.ai/en/atlas/nanotoll.dev
- Site: https://nanotoll.dev
- For: Developers building LLM agents who want to compress context on each call and verify the result.
- Categories: agents
- Use 0/100 · alive 100/100

## Used for

- strip comments and prune JSON before model calls
- shrink long agent context with extractive token pruning
- check compression fidelity through per-call signed receipts

## Price

Not read yet.

## For developers

- MCP server: dev.nanotoll/nanotoll-mcp (namespace; open on 2026-10-26, 1 tools)
- Read 2026-10-26

## MCP server, measured daily

Server: https://mcp.nanotoll.dev/mcp, checked every day since 2026-10-26 (1 checks).

- Answered: measuring, 1 daily checks so far
- Median first answer: 673 ms
- Sign-in: none
- Context its tool definitions take: about 358 tokens (1 tools)
- Tool definitions: A, 100/100
- Method: https://inite.ai/en/atlas/mcp/report

## Your data

- Certifications stated: none
- Read 2026-10-26

---

Source: INITE Atlas, https://inite.ai/en/atlas/nanotoll.dev. JSON: https://inite.ai/api/atlas/nanotoll.dev. MCP: https://inite.ai/api/mcp (atlas_lookup).