Developer Index

Firecrawl Developer is an index built for coding agents. It covers GitHub issues, merged pull requests, repository READMEs, and curated documentation sites, so an agent can answer a question about code behavior, a library or framework, an API contract, an error message, or a known bug from primary sources rather than from a general web page.

Find the issue or pull request where a bug was reported and fixed
Read the passages of a README or a documentation page that answer one specific question
Trace an API contract back to the pull request that changed it
Recover the discussion behind an error message

To give your agent access to the Developer Index, we strongly recommend using our CLI or MCP, combined with our dedicated developer skill, which you can install with:

npx skills add firecrawl/skills@firecrawl-developer-index

Endpoints

Task	Endpoint
Search the developer index	`GET` or `POST /search/developer`
Add developer results to a web search	`POST /search` with `categories: ["developer"]`

Search the developer index

Send a natural-language question and get back ranked developer results with the passages that matched. This is the path to reach for when you want developer sources only, with the result type, repository, and documentation source filters available. A developer search costs 2 credits per 10 results, rounded up (1–10 results = 2 credits, 11–20 = 4 credits, and so on). No API key is needed to get started; send one for higher rate limits.

# No API key needed to get started; add -H "Authorization: Bearer $FIRECRAWL_API_KEY" for higher rate limits:
curl -s "https://api.firecrawl.dev/v2/search/developer?query=how%20do%20I%20configure%20retries&k=10"

firecrawl developer "how do I configure retries" --limit 10

POST is available on the same path, and is the easier form when you want to pass array filters as JSON:

cURL

# No API key needed to get started; add -H "Authorization: Bearer $FIRECRAWL_API_KEY" for higher rate limits:
curl -X POST https://api.firecrawl.dev/v2/search/developer \
  -H "Content-Type: application/json" \
  -d '{
    "query": "how do I configure retries",
    "k": 10,
    "types": ["issue", "pull_request"]
  }'

Each result carries a stable id such as issue:owner/repo#123, a type of doc, issue, pull_request, or readme, a url, and its matched passages in markdown, so tables and code blocks survive. title is frequently absent on doc results, where the source page carries no usable title, so fall back to url rather than assuming the field is present. Alongside the results, coverage reports the outcome for each result type and reranked reports whether the ranked list went through the reranking stage. Check coverage when a result type you expected is missing: skipped means your types value did not ask for that type, while degraded or unavailable means the gap came from the index or from a filter, not from the query. Optional filters narrow the search:

k sets how many results come back, defaulting to 10, and passages how many matched passages each one carries
types picks which of doc, issue, pull_request, and readme to search
repos scopes the GitHub half of the index, and sources scopes the documentation half
skills set to only limits the search to indexed agent-skill files
language, topic, license, min_stars, max_stars, archived, and fork filter on repository attributes, such as language=Rust, topic=async, or license=MIT

Those seven repository filters describe a GitHub repository, so sending one without a sources scope returns no doc results and reports doc as unavailable in coverage. Read how the repository filters scope a search before you send one. See the developer search reference for every filter’s type and bounds, how repos and sources scope a search, and the full response schema.

The Python and Node SDKs reach the Developer Index through the developer category shown below. They do not expose a dedicated method for this endpoint, so call it over HTTP, through the CLI, or through MCP.

Add developer results to a web search

Pass developer in the categories array on /search when you are already calling /search and want developer results weighed alongside ordinary web results in a single call. The API returns them in a developer group next to web, and both SDKs expose that group as .developer. No API key is needed to get started — /search accepts keyless requests, and the developer category comes with it, subject to the keyless allowance. Send a key for higher rate limits.

# No API key needed to get started; add -H "Authorization: Bearer $FIRECRAWL_API_KEY" for higher rate limits:
curl -X POST https://api.firecrawl.dev/v2/search \
  -H "Content-Type: application/json" \
  -d '{
    "query": "how do I configure retries",
    "categories": ["developer"],
    "limit": 10
  }'

firecrawl search "how do I configure retries" --categories developer --limit 10

from firecrawl import Firecrawl

firecrawl = Firecrawl(api_key="fc-YOUR-API-KEY")

result = firecrawl.search(
    "how do I configure retries",
    categories=["developer"],
    limit=10,
)
for item in result.developer or []:
    print(item.url, item.title)

import { Firecrawl } from "firecrawl";

const firecrawl = new Firecrawl({
  // No API key needed to get started; add one for higher rate limits:
  // apiKey: "fc-YOUR-API-KEY",
});

const result = await firecrawl.search("how do I configure retries", {
  categories: ["developer"],
  limit: 10,
});
for (const item of result.developer ?? []) {
  console.log(item.url, item.title);
}

Developer results in this group carry url, title, description, and position, the same shape as a web result, plus category: "developer". Web results in the same response have no category, so it is the field to key on if you merge the two groups. Results arrive in their own group rather than in web, so SDK users read them with result.developer. This surface returns the web result shape, not the ranked developer shape. For the matched passages and the index filters, use the developer search endpoint.

The hosted MCP server exposes both surfaces, and neither writes anything. See MCP tools for firecrawl_developer_search, for developer results through firecrawl_search, and for which of the two the keyless tool surface carries.

Get Started

Core Endpoints

More

Quickstarts

Developer Guides

Webhooks

Use Cases

Other

Contributing

Endpoints

Search the developer index

Add developer results to a web search

​Endpoints

​Search the developer index

​Add developer results to a web search

Endpoints

Search the developer index

Add developer results to a web search