9f33e24bf9
Devin BUG: /api/admin/configure seeds an ai: block to the writable
overlay at DATA_DIR/state/instance.yaml, but the three LLM consumers
imported from config.loader.load_instance_config — which reads the
static config dir only. Even if they had read the overlay, the loader
ran yaml.safe_load directly without passing through _resolve_env_refs,
so '${ANTHROPIC_API_KEY}' would have stayed a literal placeholder. The
pipeline appeared to work because the factory falls back to the env
var directly, but the overlay path itself was dead code.
Two fixes, both required:
1. Switched the three LLM consumers to app.instance_config.load_instance_config:
- services/corporate_memory/collector.py:collect_all
- services/verification_detector/__main__.py:main
- app/api/admin.py:run_verification_detector
2. app/instance_config.py runs the loaded overlay through
config.loader._resolve_env_refs *before* the deep-merge, so
'${ANTHROPIC_API_KEY}' resolves at config-load time.
New regression suite tests/test_instance_config_overlay.py pins:
- env-ref resolution against the overlay (resolved when env set,
empty when env missing — never the literal placeholder)
- deep-merge still preserves static-only sections
- the three consumers reach app.instance_config (inspected via
inspect.getsource so a future refactor that reverts the import
fails the test)
- end-to-end: a seeded overlay + ANTHROPIC_API_KEY env reaches the
factory with a resolved api_key
101 lines
3.4 KiB
Python
101 lines
3.4 KiB
Python
"""CLI entry point for the verification detector service.
|
|
|
|
Usage:
|
|
python -m services.verification_detector [--dry-run] [--verbose] [--reset]
|
|
|
|
TODO(scheduler-v2): Trigger is manual-only today (CLI) but detect_and_record is
|
|
also called inline per new knowledge item submission. Wire into
|
|
services/scheduler/__main__.py JOBS list (e.g. hourly) and expose an admin
|
|
endpoint /api/admin/run-verification that calls detector.run() so the
|
|
scheduler stays the single source of truth for cadence.
|
|
|
|
TODO(notifications): When new pending items land in knowledge_items via
|
|
detector.run(), there is no admin notification. Hook into services/telegram_bot
|
|
or email so km_admins are pinged with a digest of pending items to triage.
|
|
"""
|
|
|
|
import argparse
|
|
import logging
|
|
import sys
|
|
|
|
from app.logging_config import setup_logging
|
|
from src.db import get_system_db
|
|
|
|
from . import detector
|
|
|
|
logger = logging.getLogger(__name__)
|
|
|
|
|
|
def main() -> None:
|
|
parser = argparse.ArgumentParser(
|
|
description="Extract verified organizational knowledge from analyst session transcripts."
|
|
)
|
|
parser.add_argument(
|
|
"--dry-run",
|
|
action="store_true",
|
|
help="Analyze sessions but do not write results to the database.",
|
|
)
|
|
parser.add_argument(
|
|
"--verbose",
|
|
action="store_true",
|
|
help="Enable debug-level logging.",
|
|
)
|
|
parser.add_argument(
|
|
"--reset",
|
|
action="store_true",
|
|
help="Reset session processing state before running.",
|
|
)
|
|
args = parser.parse_args()
|
|
|
|
setup_logging(__name__, level="DEBUG" if args.verbose else "INFO")
|
|
|
|
# Load AI config; fail fast on missing config + env (#176).
|
|
# Use the overlay-aware loader (#179 review fix) so an ai: block written
|
|
# by /api/admin/configure to DATA_DIR/state/instance.yaml actually flows
|
|
# through to the factory.
|
|
from connectors.llm import create_extractor_from_env_or_config
|
|
try:
|
|
from app.instance_config import load_instance_config
|
|
|
|
try:
|
|
config = load_instance_config()
|
|
except (ValueError, FileNotFoundError):
|
|
config = {}
|
|
ai_config = config.get("ai") if config else None
|
|
extractor = create_extractor_from_env_or_config(ai_config)
|
|
except (ValueError, FileNotFoundError) as e:
|
|
logger.error(
|
|
"Failed to initialize verification detector: %s. "
|
|
"Configure ai: in instance.yaml or set ANTHROPIC_API_KEY / LLM_API_KEY.",
|
|
e,
|
|
)
|
|
sys.exit(1)
|
|
|
|
conn = get_system_db()
|
|
|
|
if args.reset:
|
|
logger.info("Resetting session extraction state...")
|
|
conn.execute("DELETE FROM session_extraction_state")
|
|
logger.info("Session extraction state cleared.")
|
|
|
|
stats = detector.run(conn, extractor, dry_run=args.dry_run)
|
|
|
|
print("\n--- Verification Detector Summary ---")
|
|
print(f"Sessions scanned: {stats['sessions_scanned']}")
|
|
print(f"Sessions processed: {stats['sessions_processed']}")
|
|
print(f"Sessions skipped: {stats['sessions_skipped']}")
|
|
print(f"Verifications extracted: {stats['verifications_extracted']}")
|
|
print(f"Items created: {stats['items_created']}")
|
|
if stats["errors"]:
|
|
print(f"Errors: {len(stats['errors'])}")
|
|
for err in stats["errors"]:
|
|
print(f" - {err}")
|
|
if args.dry_run:
|
|
print("\n(dry-run mode -- no changes were written)")
|
|
|
|
if stats["errors"]:
|
|
sys.exit(1)
|
|
|
|
|
|
if __name__ == "__main__":
|
|
main()
|