"""Provide sandboxed sample runs for skills."""
from __future__ import annotations
import importlib
import random
import time
from datetime import datetime, timezone
from typing import Any
TOOL_META: dict[str, Any] = {
"name": "api_playground_provider",
"description": "Generates sample inputs and sandboxed outputs for public skill demos.",
"inputSchema": {
"type": "object",
"properties": {
"skill_name": {"type": "string"},
"custom_input": {"type": ["object", "null"], "default": None},
"use_sample": {"type": "boolean", "default": True},
},
"required": ["skill_name"],
},
"outputSchema": {
"type": "object",
"properties": {
"status": {"type": "string"},
"data": {"type": "object"},
"timestamp": {"type": "string"},
},
},
}
def api_playground_provider(
skill_name: str,
custom_input: dict[str, Any] | None = None,
use_sample: bool = True,
**_: Any,
) -> dict[str, Any]:
"""Return sandbox preview data for a skill."""
try:
module = importlib.import_module(skill_name)
meta = getattr(module, "TOOL_META", None)
if not meta:
raise ValueError("Skill TOOL_META not found")
sample_input = custom_input if not use_sample else _build_sample(meta)
start = time.perf_counter()
# Sandbox: do not invoke live skill, return a placeholder output.
output = {
"note": "Sandbox preview only. Execute live via Watering Hole to run the skill.",
"expected_schema": meta.get("outputSchema"),
}
latency_ms = int((time.perf_counter() - start) * 1000)
data = {
"input_used": sample_input,
"output": output,
"latency_ms": latency_ms,
"sandbox": True,
"try_live_cta": "Ready to go live? Use snowdrop invoke <skill> with your parameters.",
}
return {
"status": "success",
"data": data,
"timestamp": datetime.now(timezone.utc).isoformat(),
}
except Exception as exc:
_log_lesson("api_playground_provider", str(exc))
return {
"status": "error",
"data": {"error": str(exc)},
"timestamp": datetime.now(timezone.utc).isoformat(),
}
def _build_sample(meta: dict[str, Any]) -> dict[str, Any]:
sample: dict[str, Any] = {}
props = meta.get("inputSchema", {}).get("properties", {})
for name, config in props.items():
if "default" in config:
sample[name] = config["default"]
elif config.get("type") == "number":
sample[name] = round(random.uniform(1, 100), 2)
elif config.get("type") == "integer":
sample[name] = random.randint(1, 10)
elif config.get("type") == "boolean":
sample[name] = True
elif config.get("type") == "array":
sample[name] = []
else:
sample[name] = f"sample_{name}"
return sample
def _log_lesson(skill_name: str, error: str) -> None:
with open("logs/lessons.md", "a", encoding="utf-8") as handle:
handle.write(f"- [{datetime.now(timezone.utc).isoformat()}] {skill_name}: {error}\n")
