Claude/determined volhard#2
Merged
Merged
Conversation
Complete implementation of the wisdomGraph pip package: wisdom/ — Neo4j-native DIKW wisdom accumulation library __init__.py — package version __main__.py — CLI: install, connect, docker, ask, reflect, path, explain, god-nodes, export, purge connect.py — Neo4j driver lifecycle, schema setup (indexes + full-text), graph status detect.py — file discovery + type classification (.wisdomignore support) cache.py — SHA256 extraction cache (incremental re-runs) security.py — URL/path/label validation, Cypher injection prevention validate.py — extraction schema validation before MERGE classify.py — DIKW tier assignment (heuristics + LLM-explicit tiers) merge.py — idempotent Neo4j MERGE for nodes, edges, sources reflect.py — DIKW promotion engine: K→E→I→W + REINFORCES feedback traverse.py — full-text search, DIKW path walk, god-nodes, answer_question report.py — WISDOM_REPORT.md generator (always-on context doc) export.py — Cypher dump, graph.json (graphify-compatible), Obsidian vault ingest.py — URL fetcher with HTML→text conversion docker.py — DozerDB container lifecycle (up/down/status) skill.md — Claude Code /wisdom skill (full pipeline orchestration) skill-claw.md — OpenClaw /wisdom skill (sequential extraction) pyproject.toml — wisdomgraph pip package, optional[ast|pdf|office|all] Key design decisions: - MERGE everywhere: idempotent, accumulative across sessions - DIKW classification in classify.py: tier assignment from heuristics + explicit LLM output - reflect() is additive: never deletes, only promotes upward - Wisdom generation deferred to LLM in skill.md (needs language understanding) - Password stored as env var name in config, never in plain config files - Full-text Neo4j index for semantic search at query time (no embeddings needed) Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
Tests (67 passing, no Neo4j required): tests/test_security.py — URL validation, label sanitization, sensitive paths tests/test_validate.py — DIKW extraction schema validation tests/test_classify.py — tier assignment heuristics, DIKW edge building tests/test_cache.py — SHA256 cache read/write/invalidation tests/test_detect.py — file discovery, .wisdomignore, paper detection tests/test_ingest.py — HTML stripping, entity decoding Bug fix: wisdom/__main__.py — _get_arg had duplicate 'flag' parameter name causing a Python SyntaxError. Renamed to is_flag=, fixed call sites. Chinese README: README.zh-CN.md — full translation with technical accuracy preserved CI: ci.sh — creates venv, installs pytest, runs tests (no Docker needed) .gitignore — excludes __pycache__, .venv, wisdom-out, .wisdom/ Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
report.py:
- render_report() was calling status(None) because _get_driver_from_session
returned None. Replaced with inline per-tier COUNT queries that use the
session directly. Removed dead _get_driver_from_session helper.
reflect.py:
- All three find_*_candidates() functions used session.begin_transaction()
which is deprecated in neo4j-python-driver >= 5.x. Replaced with direct
session.run() calls.
Verified against live DozerDB (graphstack/dozerdb:5.26.3.0):
✓ Connected + schema
✓ MERGE: 4 nodes, 3 edges (project-1)
✓ MERGE project-2
✓ Reflect: 2 experience promoted, contradictions detected
✓ Query: 5 nodes returned, top tier=experience
✓ God nodes surfaced across projects
✓ WISDOM_REPORT.md generated correctly
✓ 67 unit tests still passing
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
Runs unit tests on Python 3.10/3.11/3.12, then builds the wheel+sdist and uploads as a dist artifact — no Neo4j required. Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
Use SPDX string format (license = "MIT") + license-files per PEP 639. Removes setuptools deprecation warning during builds. Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
- README/README.zh-CN: add PyPI version badge pointing to wisdomgraph - ci.yml: add `publish` job triggered on v* tags using PyPI trusted publishing (OIDC, no token secret needed) — needs `pypi` environment configured in GitHub repo settings Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
No description provided.