From generation tool to knowledge system
v3.0.0 is a system-level redesign of VaultForge — transforming it from a text generator into an engineering support system for cognitive processes. Three pillars: runtime reliability, incremental knowledge accumulation, and output precision control.
AI agent skill for Obsidian. Give it a PDF, get a complete vault — atomic notes, logical wikilinks, MOCs, and deep research. 58 automated tests. 7-phase pipeline. ⭐ Star on GitHub · Dev.to
State machine + atomic writes ensure crash recovery. Interrupted mid-work? Restart and it resumes where it stopped.
Phase 0 detects existing notes. Add new materials without touching what you've edited. Three-tier ACL protection.
5-type logical wikilinks + 6-dimension quality review. Three-stage funnel balances precision and cost.
Agent crash or network interruption → all progress lost. Files may be corrupted. Restart from page one.
Every note carries a status field. The filling state splits into two crash sub-scenarios, recovered separately. .md.tmp → verify → rename guarantees file integrity.
Read all files → generate roadmap → create all notes → build all links. New files trigger full re-run. Existing notes overwritten or orphaned.
Phase 0 scans existing notes before any content generation. Pristine = refreshable. User-modified = protected. Locked = frozen. Incremental mode processes only new files; existing notes untouched.
| Status | Detection | Behavior |
|---|---|---|
| Pristine | Unmodified since generation | Auto-refreshable |
| User-modified | mtime differs from date | Protected from auto-writes |
| Locked | Manually set by user | Completely ignored |
New ↔ New: links written directly. New → Existing: suggestions JSON only, existing files untouched. The link network in existing notes is the user's mental model — new links should be an optional data source, not an automatic change.
Two notes sharing a keyword get linked. Stops at recognition memory — "I've seen this word" rather than "how do these concepts relate."
Derivation · Analogy · Contradiction · Application · Context — five types covering Bloom's taxonomy from comprehension to evaluation. note‑reviewer independently audits every note across six dimensions.
| Relation | Meaning | Bloom's Level |
|---|---|---|
| Derivation | A is the logical prerequisite of B | Understand → Apply |
| Analogy | A and B share deep structure but surface differently | Analyze |
| Contradiction | A and B reach conflicting conclusions | Evaluate |
| Application | A's framework is used in B's context | Apply |
| Context | A and B are phases in the same domain's evolution | Synthesize |
VaultForge v3.1.0 is the most educationally rigorous and architecturally disciplined project in the current Agent Skill ecosystem. It has evolved from a one-shot generation tool into a continuous knowledge management system — with Phase 0 metacognitive awareness, vf_ frontmatter as a persistent state layer, ACL-style three-tier note protection, and engineering quality backed by 58 tests and 16/16 audit closure.
| Pedagogical Design | v1.2.0 | v3.1.0 | Driver |
|---|---|---|---|
| Cognitive Load Management | 3.5 | 5.0 | Atomic note proportion check + Context Pre-extraction |
| Advance Organizers | 4.0 | 5.0 | MOC wikilink completeness auto-audit |
| Semantic Network | 3.0 | 4.5 | Five-type classification replaces keyword matching |
| Retrieval Guidance | 3.0 | 4.0 | Structured guidance replaces simple lists |
| Critical Thinking | 3.0 | 5.0 | Three-layer Controversy Analysis framework |
| Spiral Curriculum | — | 5.0 | Phase 0 + Incremental + three-tier ACL |
| Learner Autonomy | — | 4.5 | vf_status three-tier ACL model |
| Engineering & Agent Design | v1.2.0 | v3.1.0 | Driver |
|---|---|---|---|
| Fault Tolerance | 1.0 | 5.0 | State machine + Atomic Write + 8 degradation paths |
| Output Accuracy | 2.5 | 4.5 | 6-D Review + funnel wikilinks + file-based stats |
| Token Efficiency | 2.0 | 4.5 | Context Pre-extraction + --note-filter |
| Cross‑Platform | 2.0 | 4.5 | Tri-mode CLI + Phase 0 no sub-agent dependency |
| Maintainability | 2.0 | 5.0 | 58 tests + 7 cross-file contracts + 16/16 audit |
| Internal Consistency | 2.0 | 5.0 | 19 cross-checks + automated guardrails |
| Incremental Architecture | — | 5.0 | Three-tier ACL + add-only + suggestion isolation + refresh loop |
git clone https://github.com/Easonnotsing/VaultForge.git ~/.agents/skills/VaultForge pip install pypdf
Trigger in Claude Code, Codex, or Cursor. Select output language first. Phase 0 scans automatically — if a knowledge base exists, you'll be asked: incremental update or fresh generation.