Buckets:

download
raw
2.95 kB
#!/usr/bin/env python3
"""Report effGen's headline counts (tools, prompt templates, presets, providers,
catalogued models) straight from the live package surfaces.
These numbers drift as the framework grows, so documentation and badges should be
populated from this script rather than hand-maintained. It is the single source of
truth for "how many X does effGen ship".
python scripts/gen_counts.py # human-readable table
python scripts/gen_counts.py --json # machine-readable JSON
The model count reflects the bundled offline catalog snapshot (run
``effgen models refresh`` to update it); every other count is computed by
introspecting the installed package, so it is always exact.
"""
from __future__ import annotations
import argparse
import json
import sys
def collect() -> dict:
counts: dict[str, object] = {}
# Builtin tools — the registry lazily discovers them on first access.
from effgen.tools import get_registry
counts["tools"] = len(get_registry().list_tools())
# Two distinct prompt surfaces:
# - the default TemplateManager (generic, reusable templates)
# - the domain prompt library (curated, evaluated prompts; `effgen prompts list`)
from effgen.prompts import create_default_template_manager
counts["prompt_templates"] = len(create_default_template_manager().list_templates())
from effgen.prompts.library import registry as prompt_library
counts["prompt_library"] = len(prompt_library)
# Ready-to-use agent presets.
from effgen import list_presets
counts["presets"] = len(list_presets())
# Providers and their catalogued models (offline snapshot).
from effgen.models import registry as model_registry
providers = sorted(model_registry.list_providers())
per_provider = {p: len(model_registry.list_models(p)) for p in providers}
counts["providers"] = len(providers)
counts["models"] = sum(per_provider.values())
counts["models_by_provider"] = per_provider
return counts
def main() -> int:
ap = argparse.ArgumentParser(description=__doc__)
ap.add_argument("--json", action="store_true", help="emit JSON")
args = ap.parse_args()
counts = collect()
if args.json:
print(json.dumps(counts, indent=2, sort_keys=True))
return 0
print("effGen package counts")
print("=" * 40)
print(f" Tools {counts['tools']:>5}")
print(f" Prompt templates {counts['prompt_templates']:>5}")
print(f" Prompt library {counts['prompt_library']:>5}")
print(f" Presets {counts['presets']:>5}")
print(f" Providers {counts['providers']:>5}")
print(f" Models (catalog) {counts['models']:>5}")
print("-" * 40)
print(" Models by provider:")
for prov, n in sorted(counts["models_by_provider"].items()):
print(f" {prov:<12} {n:>5}")
return 0
if __name__ == "__main__":
sys.exit(main())

Xet Storage Details

Size:
2.95 kB
·
Xet hash:
2b457d54119621d248c20717323ed1a156d2103f31f04e22574f775440288b13

Xet efficiently stores files, intelligently splitting them into unique chunks and accelerating uploads and downloads. More info.