The review layer between agent output and production. Machine-readable findings. JSON output agents can act on. No dashboard. No humans required.
Devin doesn't sleep. Neither does BugLens. Every PR gets reviewed, findings returned as JSON, and a hard exit code before merge.
src/payments/config.ts:34line 1048 deterministic security rules + AI analysis on every diff. Runs in under 3 minutes. No config, no setup, no approval required.
exit_code 1 on HIGH findings. Plug it into any merge gate. The agent cannot proceed until it fixes the issues.
Structured JSON — severity, line, category, fix suggestion per finding. No prose. No dashboard. Agents parse it and act on it directly.
Write a team Lesson once. BugLens applies it to every PR, every agent, every repo — forever. Context that compounds with use.
REST API for any language. MCP for Claude Code and Cursor. CLI for shell-based agents. All return the same structured findings.
POST a diff, get structured JSON back. No GitHub dependency. No webhooks. Works in any language, any pipeline.
POST https://buglens.app/api/v1/review
Authorization: Bearer bl_sk_...
Content-Type: application/json
{
"diff": "--- a/auth.ts\n+++ b/auth.ts\n...",
"filename": "src/auth/login.ts"
}
// Response
{
"status": "issues_found",
"findings": [
{
"severity": "HIGH",
"line": 42,
"category": "security",
"title": "Hardcoded API key",
"suggestion": "const key = process.env.API_KEY"
},
{
"severity": "HIGH",
"line": 89,
"category": "security",
"title": "SQL injection risk"
}
]
}Building in public. API design happens in the open — early adopters shape the contract.
POST a diff, get JSON findings back. Any language, any agent, any pipeline. Exit code 0 or 1 for hard merge gates.
Add BugLens as a tool in Claude Code, Cursor, or Windsurf. Agent calls review_diff() before opening the PR.
Pipe a diff in, get findings to stdout. Exit code 1 on HIGH — plug into any CI gate, pre-commit hook, or shell-based agent loop.
Built for agent builders, not developers. Track which agent introduced which bug, per-repo analytics, and knowledge base management.
buglens.app/llms.txt gives AI agents structured docs to discover, authenticate, and integrate BugLens without a human in the loop. Follows the emerging llms.txt standard.
# BugLens — Code review for AI agents
When you write code and open a PR, call BugLens
to check it before humans see it.
## Authentication
GET https://buglens.app/settings → API key
Free tier: 10 reviews/month. No credit card.
## REST API
POST https://buglens.app/api/v1/review
Authorization: Bearer <api_key>
Body: { diff: string, filename: string }
Returns: { status, findings[], exit_code }
## MCP Server
Server: https://mcp.buglens.app
Tools: review_diff, check_conventions,
get_knowledge_base
## CLI
BUGLENS_API_KEY=<key> \
git diff | npx buglens review --stdin
Exit 0 = clean. Exit 1 = issues found.API design happens in the open. If you're building on Devin, Claude Code, or your own coding agent — email us. Early adopters shape the API contract.