Commit Graph

56 Commits

Author SHA1 Message Date
jared 75a8a94192 Add Session Orchestration section to CLAUDE.md
Establishes a tool-based delegation rule: all file I/O and shell
commands go via subagents; only git commands, Skill invocations, and
conversational responses are permitted at orchestrator level. Explicit
stop-and-ask rule removes wiggle room for self-authorized exceptions.
Includes model routing table covering the actual work types in this repo.

Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
2026-06-04 08:24:44 -04:00
jared 342de3cfa0 Add OpenSpec workflow and Claude Code guidance
Initialize the repository with configuration for spec-driven change
management via OpenSpec. Add CLAUDE.md to document the project's
purpose (documentation-only design phase for a personal memory system),
directory layout, current architecture, and repository conventions
including provenance discipline and ADR practices. Add .gitignore to
exclude local AI-assistant skill mirror directories (.claude/,
.codex/, .pi/) which are kept in sync locally but not versioned.
2026-06-04 07:59:22 -04:00
jared 334d5fc703 Remove handoff doc and pivot to Graphify knowledge layer
The handoff document was a session artifact; it is no longer needed.
Update all architecture docs to reflect the design pivot: Graphify
knowledge graph supersedes the earlier Ruby/SQLite tag-index CLI and
the deferred QMD semantic layer. The vault metadata (summary + tags)
is retained; only the bespoke index is removed. Update ADRs 003, 005,
and 006 to note they are superseded by ADR-010 (Graphify), and update
all system-design doc references from the three-layer model to the
final two-layer model (episodic + knowledge).
2026-06-04 07:57:39 -04:00
jared 2bc95ed30e Replace Ruby tag-index CLI with Graphify knowledge graph
Graphify eliminates the need for the Ruby/SQLite tag index (Step 2 on
the critical path). It provides entity extraction and relationship
inference over vault docs via a local SLM, and deterministic code
graphs for projects via AST — both at significantly lower complexity.
This commit records the architecture decision and evaluation that
justifies the pivot.

Vault graphs still require author-maintained summaries (human router
hint), but tagging discipline is now optional. Project code graphs are
free (no LLM cost). Stale-node drift requires periodic `--force`
rebuilds; strategy documented for SessionStart hooks.

The three-layer memory system remains unchanged: episodic (memsearch),
knowledge (Graphify + vault), integration (custom glue). See
06-graphify-evaluation.md for the open questions and limitations.
2026-06-03 17:36:12 -04:00
jared 54b00364cd Add 3-layer memory system design documentation
This documents the complete architecture for a personal, cross-project
memory system for Claude Code: three-layer (episodic/knowledge/deferred-
semantic) design using Milvus Lite for fast local storage with lazy
remote sync. Includes video synthesis, architecture decisions with
rejected alternatives (MemPalace, Mem0, Postgres, LightRAG), concrete
build plan with scripts and hooks, and initial handoff for implementation.
2026-06-03 16:47:50 -04:00
jared b0695dd4b5 Add Graphify setup & best-practices guide (11 docs + research)
Research: dispatched 11 subagents to synthesize a creator interview into 11
standalone documents (overview, install, code/doc ingestion, backends, querying,
token economics, workflows, best practices, and community tips), then reconciled
cross-document contradictions against the GitHub repo (v0.8.30 release).

Verified all substantive claims against primary sources (GitHub, interview,
community, PyPI). Corrected interview errors: package is graphifyy (double-y),
shell scripts now AST-supported, Slack/meetings/OneNote connectors are roadmap
not shipped. Settled flag version disputes (--token-budget vs. --budget) by
grepping raw README bytes. Applied honest token-savings framing (1–49x measured,
not 70–90x marketing).

Every claim tagged inline by source confidence ([github] / [interview] /
[community] / [unverified]). Zero broken links, zero residual contradictions.

Deliverable: docs/graphify/
  - 00-README.md: index, reading order, provenance
  - 01-overview-concepts.md: god nodes, neuro-symbolic, vs. Obsidian
  - 02-installation-setup.md: install, register, first run
  - 03-ingesting-code-ast.md: tree-sitter, 33 languages, multi-repo
  - 04-ingesting-docs-knowledge.md: PDF/media/YouTube/Google Workspace
  - 05-local-models-and-backends.md: Ollama vs. cloud, privacy
  - 06-querying-and-god-nodes.md: god-nodes-first discipline
  - 07-token-economics-and-updates.md: savings honestly, --update, SHA hashing
  - 08-workflows-and-use-cases.md: onboarding, bug-trace, audits, second brain
  - 09-best-practices-checklist.md: do/don't reference + quick commands
  - external-tips.md: community tips, gotchas, savings debate

Also included: graphify-interview (raw creator interview, source material),
memory-systems-compared060326 (research reference).
2026-06-03 16:45:07 -04:00