Spaces:
Configuration error
Download docs/api.md from varshithreddy1807/aria: direct link, hf CLI and curl.
- Browser
- Download file 2.68 kB
-
https://huggingface.co/spaces/varshithreddy1807/aria/resolve/main/docs/api.md
- Command line
-
hf download hf://spaces/varshithreddy1807/aria/docs/api.md
-
curl -L -o api.md https://huggingface.co/spaces/varshithreddy1807/aria/resolve/main/docs/api.md
API Guide — ARIA v1.5.0
The ARIA API is built on FastAPI and operates within the API Layer of the Repository Intelligence Architecture (RIA). It exposes HTTP REST endpoints, server-sent events (SSE) progress streams, and Prometheus metrics.
All endpoints are versioned under /api/v1/ while maintaining legacy root paths (/api/...) for 100% backward compatibility.
Observability & Metrics
GET /metrics or GET /api/v1/metrics
Exposes application health metrics, active requests, build durations, individual analysis durations, and cache hit/miss stats.
Sample Scrape Response:
# HELP http_requests_total Total number of HTTP requests.
# TYPE http_requests_total counter
http_requests_total{method="GET",path="/health",status="200"} 12.0
# HELP active_requests_count Total number of active requests.
# TYPE active_requests_count gauge
active_requests_count 0.0
# HELP cache_hits_total Total number of analysis cache hits.
# TYPE cache_hits_total counter
cache_hits_total{cache_key="symbols"} 3.0
Key REST Endpoints
1. POST /api/v1/analyze or POST /api/analyze
Triggers code checkout, syntax parsing, and AI embeddings indexing. Streams SSE progress JSON.
- Request Body:
{ "url": "https://github.com/owner/repo", "branch": "main", "model": "deepseek-ai/deepseek-v4-flash" }
2. GET /api/v1/architecture/{owner}/{repo}/graph or GET /api/architecture/{owner}/{repo}/graph
Exposes the repository dependency graph in a React-Flow-friendly node/edge schema.
- Response:
{ "nodes": [ { "id": "main.py", "type": "file" } ], "edges": [ { "id": "e1", "source": "main.py", "target": "utils.py" } ] }
3. GET /health or GET /api/v1/health
Returns detailed application settings validation, LLM providers, and storage path health check details.
4. POST /api/v1/report/{owner}/{repo}/build
Generates or updates the Repository Intelligence Report for the specified repository and returns the complete report model JSON.
- Response:
ReportDataModel
5. GET /api/v1/report/{owner}/{repo}/summary
Fetches a brief summary of the latest health report (overall score, grade, generated timestamp).
6. GET /api/v1/report/{owner}/{repo}/download
Downloads the compiled repository intelligence report.
- Query Parameters:
format(optional):html(default),pdf, ormarkdown.
- Response: Binary stream with
Content-Disposition: attachment; filename="..."header.
7. GET /api/v1/repositories/{owner}/{repo}/workspace
consolidated snaphot endpoint for all IDE panel datasets.