Skip to main content

Documentation Index

Fetch the complete documentation index at: https://docs.usedino.dev/llms.txt

Use this file to discover all available pages before exploring further.

Yes. Configure apis[].type: rest with source: openapi and a specPath in tenants/<id>.yml, then run dino scan --tenant <id>. REST shares the same pipeline, catalogue, and reporting model as GraphQL. See Quick Start and Configuration.
Only to reach your API endpoint (and, for REST, to fetch specPath when it is a URL). The CLI runs entirely locally. If you enable AI reasoning, it calls the Anthropic API (requires DINO_AI_KEY). Without AI features, Dino makes zero external calls beyond your API and any remote OpenAPI URL you configure.
No. Dino runs on your machine. API responses are processed locally. The only external call is the optional AI reasoning feature, which strips all secrets before sending data to the LLM provider.
Yes. Every command supports --quiet and returns meaningful exit codes:
  • dino scan --fail-on-high — exit 1 on HIGH or CRITICAL findings
  • dino diff --fail-on-breaking — exit 1 on breaking changes
  • dino lint --fail-on-undocumented — exit 1 on doc regressions
  • dino watch --once --autonomy enforce — single run, fail on issues
See dino watch for continuous runs in CI pipelines.
Dino is an autonomous API Quality Intelligence Platform, not a static linter. It runs deterministic quality modules (fuzzing, auth boundary testing, rate limiting, response validation, error safety, deprecation tracking, and rest-fuzzer for REST) and builds a complete health picture. Spectral checks static rules. GraphQL Inspector diffs schemas. Dino exercises live behavior.
Individual module failures don’t crash the pipeline. Failed modules are reported with error details. The run continues with remaining modules. A run where all modules fail exits 1 with no report — it does not produce a false CLEAN result.
The CLI and MCP server are MIT licensed. The cloud dashboard and enterprise features (SSO, audit logs, BYOI) are paid.
3,000+ across 420 test suites. We test for false output (wrong-but-clean results) as our #1 bug class — it’s treated as a P0 incident. See Deterministic Engine.
Yes. Use --tools to select which modules run:
dino scan --tools input-fuzzer,response-validator,rest-fuzzer
By default, all shipped modules run. RBAC Matrix auto-skips if auth isn’t configured.
Node.js 22 or later. Dino uses modern Node APIs (native fetch, structured clone). Check with node --version.