Files
cleveragents-core/benchmarks/skill_registry_persist_bench.py
freemo 7235d46ade
CI / quality (pull_request) Successful in 19s
CI / lint (pull_request) Successful in 21s
CI / benchmark-publish (pull_request) Has been skipped
CI / security (pull_request) Successful in 50s
CI / typecheck (pull_request) Successful in 58s
CI / build (pull_request) Successful in 29s
CI / integration_tests (pull_request) Successful in 4m16s
CI / unit_tests (pull_request) Successful in 12m18s
CI / docker (pull_request) Successful in 1m30s
CI / benchmark-regression (pull_request) Successful in 25m15s
CI / coverage (pull_request) Successful in 1h21m51s
CI / build (push) Successful in 15s
CI / quality (push) Successful in 17s
CI / lint (push) Successful in 21s
CI / security (push) Successful in 28s
CI / typecheck (push) Successful in 31s
CI / benchmark-regression (push) Has been skipped
CI / integration_tests (push) Successful in 2m47s
CI / unit_tests (push) Successful in 11m0s
CI / docker (push) Successful in 40s
CI / benchmark-publish (push) Successful in 12m22s
CI / coverage (push) Successful in 44m42s
feat(skill): persist flattened tool sets
Add Alembic migration m4_002_skill_flattened_tools to extend the skills
table with five new columns: flattened_tools_json, includes_json,
capability_summary_json, yaml_text, and flattening_hash (SHA-256).  A
defence-in-depth uniqueness constraint (uq_skills_name) is also added.

Update SkillModel with the new column definitions and extend
SkillRepository with update_flattened_tools(), get_flattened_tools(),
needs_refresh(), recompute_flattening_hash(), and
invalidate_cached_summaries() methods.  The existing update() method
now nulls all cached fields on mutation (hash-based invalidation).

All new repository methods follow the session-factory pattern with
@database_retry and flush-but-don-t-commit semantics.  Structured
logging via structlog records cache updates and invalidations.

Database schema docs updated with the new skills table columns and a
persistence-field-to-domain-model mapping table.

Tests:
- 6 Behave scenarios covering create, invalidation, hash staleness,
  refresh recomputation, uniqueness constraint, and namespace filtering
- 2 Robot Framework smoke tests (round-trip and invalidation)
- 3 ASV benchmarks (persist, refresh check, namespace list)

ISSUES CLOSED: #166
2026-02-27 20:09:26 +00:00

131 lines
4.2 KiB
Python

"""ASV benchmarks for Skill Registry flattened tool persistence.
Measures performance of:
- Persisting flattened tools with ``update_flattened_tools``
- Refresh-check via ``needs_refresh``
- Namespace-filtered listing via ``list_all``
"""
from __future__ import annotations
import hashlib
import json
import sys
import uuid
from pathlib import Path
try:
from sqlalchemy import create_engine
from sqlalchemy.orm import Session, sessionmaker
from cleveragents.domain.models.core.skill import Skill
from cleveragents.infrastructure.database.models import Base
from cleveragents.infrastructure.database.repositories import (
SkillRepository,
)
except ModuleNotFoundError:
sys.path.insert(0, str(Path(__file__).resolve().parents[1] / "src"))
from sqlalchemy import create_engine
from sqlalchemy.orm import Session, sessionmaker
from cleveragents.domain.models.core.skill import Skill
from cleveragents.infrastructure.database.models import Base
from cleveragents.infrastructure.database.repositories import (
SkillRepository,
)
def _make_skill(name: str) -> Skill:
return Skill(name=name, description=f"Benchmark skill {name}")
class SkillPersistWithTools:
"""Benchmark persisting flattened tools on an existing skill."""
timeout = 60.0
def setup(self) -> None:
self.engine = create_engine("sqlite:///:memory:", echo=False)
Base.metadata.create_all(self.engine)
self.factory: sessionmaker[Session] = sessionmaker(bind=self.engine)
self.repo = SkillRepository(session_factory=self.factory)
self.skill = _make_skill("bench/persist-tools")
self.repo.create(self.skill)
self.factory().commit()
self.tools_json = json.dumps([f"tool-{i}" for i in range(20)])
self.yaml_text = "name: bench/persist-tools\ntools: [...]"
self.hash_val = hashlib.sha256(self.yaml_text.encode()).hexdigest()
def teardown(self) -> None:
self.engine.dispose()
def time_skill_persist_with_tools(self) -> None:
self.repo.update_flattened_tools(
name="bench/persist-tools",
flattened_tools_json=self.tools_json,
includes_json="[]",
capability_summary_json="{}",
yaml_text=self.yaml_text,
flattening_hash=self.hash_val,
)
self.factory().commit()
class SkillRefreshCheck:
"""Benchmark refresh-check (hash comparison)."""
timeout = 60.0
def setup(self) -> None:
self.engine = create_engine("sqlite:///:memory:", echo=False)
Base.metadata.create_all(self.engine)
self.factory: sessionmaker[Session] = sessionmaker(bind=self.engine)
self.repo = SkillRepository(session_factory=self.factory)
self.skill = _make_skill("bench/refresh-check")
self.repo.create(self.skill)
self.factory().commit()
self.repo.update_flattened_tools(
name="bench/refresh-check",
flattened_tools_json="[]",
includes_json="[]",
capability_summary_json="{}",
yaml_text="name: bench/refresh-check",
flattening_hash="match_hash",
)
self.factory().commit()
def teardown(self) -> None:
self.engine.dispose()
def time_skill_refresh_check(self) -> None:
self.repo.needs_refresh("bench/refresh-check", "match_hash")
class SkillListByNamespace:
"""Benchmark listing skills filtered by namespace."""
timeout = 60.0
def setup(self) -> None:
self.engine = create_engine("sqlite:///:memory:", echo=False)
Base.metadata.create_all(self.engine)
self.factory: sessionmaker[Session] = sessionmaker(bind=self.engine)
self.repo = SkillRepository(session_factory=self.factory)
# Seed 50 skills across 5 namespaces
for ns in ("local", "devops", "data", "ml", "infra"):
for i in range(10):
uid = uuid.uuid4().hex[:8]
self.repo.create(_make_skill(f"{ns}/tool-{i}-{uid}"))
self.factory().commit()
def teardown(self) -> None:
self.engine.dispose()
def time_skill_list_by_namespace(self) -> None:
self.repo.list_all(namespace="local")