uri2run

AI Cost Tracking

• 🤖 LLM usage: $0.1754 (3 commits)
• 👤 Human dev: ~$150 (1.5h @ $100/h, 30min dedup)

Generated on 2026-06-15 using openrouter/qwen/qwen3-coder-next

---

Neutral transport runtime for capability backends.

python     — python://module:function handlers
shell      — shell commands with timeout; argv mode can avoid shell=True
http       — HTTP/HTTPS requests with headers/body/retry support (httpx)
stdio      — subprocess JSON stdin/stdout protocol
sse        — Server-Sent Events stream reader
ws         — WebSocket client (requires pip install uri2run[stream])
docker     — docker:// compose/container control via uri3 controller
ssh        — ssh:// remote resolve + command execution
mcp        — mcp:// HTTP bridge to MCP tools endpoints
a2a        — a2a:// HTTP bridge to agent-card and task endpoints
uri_flow   — expand + validate + dry-run/run compact flow
uri_graph  — validate + dry-run/run workflow graph

touri delegates python, shell, http, stdio, sse, ws, docker, ssh, mcp, a2a, uri_flow, and uri_graph backends here.

CLI

uri2run call python://touri_examples.weather:handler --payload '{"place":"Gdansk","days":14}'
uri2run call 'echo ok' --type shell
uri2run call http://localhost:8101/health

Python API

from uri2run import run_backend

result = run_backend(
    {"type": "python", "target": "python://touri_examples.weather:handler"},
    {"place": "Gdansk", "days": 14},
    {"root": "."},
)

Every result includes meta.transport, meta.duration_ms, and meta.runtime = uri2run. Backends with an explicit target also include meta.target.

Install

pip install -e .
# optional streaming extras:
pip install -e ".[stream]"

See docs/URI2RUN_ARCHITECTURE.md.

Links

• TODO
• touri
• Org status: ../TODO_STATUS.md

License

Licensed under Apache-2.0.