# Devin — API Agent Score

> Score: 69/100 (Grade: C) | Domain: devin.ai | Rubric: 1.0.0 | Checked: July 29, 2026

Devin scored 69/100 (C), classified "partially-ready".

[View full report](https://www.postman.com/ai/ai-ready-apis/company/devin)
[All organizations](https://www.postman.com/ai/ai-ready-apis/llms.txt)

## Summary

- Overall: 69/100 (C)
- Classification: partially-ready

## Category Scores

- API Design: 45/100 (F)
- Developer Experience: 82/100 (A)
- Agent Discovery: 92/100 (A+)
- Agent Understanding: 67/100 (C)
- Agent Usability: 72/100 (B)

## Check Results

### API Design

- [pass] Example coverage — README is present and includes a runnable quickstart. Investigated: sdk 100%, docs 0%.
- [fail] Schema coverage & depth — no .d.ts found. Investigated: sdk 0%.
- [fail] Security & governance hygiene — No security.txt at /.well-known/security.txt or /security.txt. Investigated: wellknown 0%.
- [skip] Machine-readable, versioned contract — No surface produced evidence for this capability in this run.
- [skip] Auth declared & discoverable — No surface produced evidence for this capability in this run.

### Developer Experience

- [pass] Self-service developer portal — Signup page at https://app.devin.ai/signup (CSR-rendered — body classifier inconclusive); free tier / sandbox documented in pricing/docs. Investigated: docs 100%.
- [pass] Quickstart present — Quickstart at https://docs.devin.ai/get-started/devin-intro has a runnable sample in the .md variant (https://docs.devin.ai/get-started/devin-intro.md). Investigated: docs 100%.
- [pass] Changelog published — Newest official SDK activity 9 day(s) ago. Investigated: sdk 100%.
- [fail] Code samples in docs — No detectable code samples across 20 sampled docs pages. Investigated: docs 0%.
- [skip] Description completeness — No surface produced evidence for this capability in this run.

### Agent Discovery

- [pass] Registry & SDK presence — Indexed on Context7 (websites/devin_ai, 5941 snippets). Investigated: docs 100%, sdk 67%, cli 0%.
- [warn] Docs reachable, not hard auth-gated — 1 of 50 sampled pages use cross-host redirects. Investigated: docs 88%.
- [warn] llms.txt present, valid & comprehensive — llms.txt contains parseable links but doesn't fully follow the proposed structure: https://docs.devin.ai/llms.txt: No blockquote summary found. Investigated: docs 81%, sdk 0%.
- [pass] Crawlable / AEO — Docs paths crawlable by all monitored AI agents. Investigated: wellknown 100%.

### Agent Understanding

- [warn] Agent-navigable, token-efficient docs — 4 of 49 pages have minor content differences between markdown and HTML. Investigated: docs 94%.
- [warn] Machine-readable errors (RFC 9457) — Single base error "null" present, but only 1 subclass found; a richer hierarchy lets consumers catch narrowly. Investigated: sdk 50%.
- [pass] Docs structured data — JSON-LD Article markup on 18/18 assessed pages (100%) with dateModified present. Investigated: docs 100%.
- [warn] Description consistency across surfaces — Mean pairwise description similarity across 2 surfaces (docs, sdk) is 4% (threshold 35% for full credit).
- [fail] Agent instructions file (AGENTS.md) — No AGENTS.md at the site root or /.well-known/. Investigated: wellknown 0%.
- [skip] Operation purpose clarity — No surface produced evidence for this capability in this run.

### Agent Usability

- [pass] Idempotency documented — Built-in retry machinery is present (grep-derived). Investigated: sdk 100%.
- [warn] Sandbox separation — No test-mode flag or sandbox environment is exposed. Investigated: sdk 50%.
- [fail] Runnable collection with test scripts — No public Postman workspace discovered for the org. Investigated: platform 0%.
- [skip] Rate-limit signaling — No surface produced evidence for this capability in this run.
- [skip] Pagination documented & consistent — No surface produced evidence for this capability in this run.

## Executive Summary

Devin's API discovery and developer onboarding story is strong, but two areas put partner program growth at risk: the API contract itself lacks the schema depth and security hygiene that partners and AI agents need to build against with confidence, and agent-facing context (instructions, runnable examples) is missing — forcing agents to guess at integration patterns rather than execute them reliably. Prioritize hardening the contract and schema first, then close the agent-context gaps so partners and their AI tools can self-serve end-to-end without friction.