The Letter Company releases beta of letterpress-1 coding agent
The Letter Company has launched the beta of Letterpress‑1, a coding agent, harness and IDE designed for growth engineering. The agent reads a website and its market signals, then builds changes in a sandbox and opens a pull request with a live preview.
Key points
- Letterpress‑1 is a coding agent that builds pull requests from site and market signals
- Pricing starts at $100/month for 10,000 tokens, with token cost 1 cent each
- The agent exposes 760 tools via MCP, REST and CLI and offers four buyer‑focused primitives
Letterpress‑1 runs on the company’s hosted platform against a user’s git repository, requiring no local GPU. Setup takes under five minutes. The product is priced in token‑based tiers: Basic $100 for 10,000 tokens, Starter $500 for 50,000, Growth $1,000 for 130,000, Scale $2,500 for 400,000, and Enterprise starting at 2 million tokens a month at half the list rate. A token equals one cent at list price, and the agent’s cost model adds 20 % to token usage, typically 150‑900 tokens per pull request ($1.50‑$9). Overage on Basic and Starter is $50 per 1,000 tokens, five times the list rate.
The agent exposes 760 tools through a single registry over MCP, REST and CLI, and offers four primitives—Proofs, Simulations, Benches and Phantoms—to attract buyers. It positions itself against tools like G2, Semrush, Profound, RB2B and Instantly, and claims to outperform Codex in a self‑reported side‑by‑side page build. No public benchmark is available, and the company has not disclosed model sizes or names.
The Letter Company Releases Letterpress-1: A Coding Agent Built for Growth Engineering
MarkTechPost · 3 November 2026
Loading the full article…
This text was published by MarkTechPost and written by Michal Sutter. It is reproduced here with attribution so you can read it in full; the rights remain with the publisher. Read it at the source ↗
The headline, key points and digest above were generated by Digest AI's editorial model from the linked sources. Automated summaries can contain errors: the sources are the record. Spotted a mistake? Tell us. Published by Martin K., who runs Digest AI and handles corrections.
More in Agents & Tools
All →- Y Combinator CEO Garry Tan says he spends up to $200,000 a year on AI tokens · 1 src
- Speedrun AI Labs joins anthropic's claude startups program · 1 src
- Nvidia and AMD back $6B AI startup Hark to challenge Grok · 2 src
- Legora’s Jacob Lauritzen explains building legal agents without seeing client data · 1 src
- OpenAI releases LifeVal evaluation for AI task performance · 1 src
Comments
via GitHub Discussions