A search index built for coding agents rather than people: 70 million READMEs, issues and pull requests, and no key needed to try it
Firecrawl launched Developer Index on 20 August: a search interface over more than 70 million developer artefacts — READMEs from public repositories, issues and pull requests from major projects, external documentation and OpenAPI specifications — refreshed continuously, most sources daily. The argument behind it is that ordinary web search was not built for machines: it matches words rather than meaning, and assembling one complete artefact used to mean stitching together dozens of API calls. On the company's own benchmark of 1,179 real developer queries, the index scores 0.63 recall at ten results, against 0.57 for Parallel, 0.54 for Mintlify and Exa, and 0.45 for plain web search. Broken down by task it reaches 0.76 on finding the right repository and 0.66 on issues and pull requests, but only 0.47 on documentation lookup. It is reachable through the API, a CLI, an MCP server and the SDKs, costs two credits per ten results, and needs no API key to start.
Related stories
Asked how many of their 'agents' actually finish a multi-step job unattended, seven in ten companies said a quarter or fewer2026-08-23The standard that plugs AI into your tools was built for a human clicking 'allow'. Its new roadmap admits the caller is now a machine.
2026-08-23A 27-billion-parameter model beat Opus 4.8 and GPT-5.5 at reproducing published papers
2026-08-23Alibaba trained a model on a rack of 100 real phones — it now finishes 92 out of 100 tasks on an actual handset
2026-08-23OpenAI has given away the engine under Codex, and its own numbers say the engine is worth more than the model
2026-08-22