By ether-moon
A knowledge distillation system that delivers only verified knowledge to AI coding agents. 3-layer architecture with convention-based air gap.
Orchestrates the Stage B distillation pipeline: discovers merged PRs labeled knowledge:pending, runs per-PR evidence collection → candidate extraction → quality gate, writes a changeset file for accepted entries, and creates a report PR for human review. Triggered on schedule (weekly/biweekly) or manual dispatch. Use when you need to process accumulated knowledge from merged PRs, run the refinement pipeline, or manually trigger a batch distillation cycle.
Collects the actual content of all evidence sources identified in a PR's Evidence Bundle Manifest and produces a structured Evidence Bundle. Stage B step 1 — transforms identifier references into full content for downstream candidate extraction. Called by batch-refine orchestrator per PR.
Processes reviewer feedback on Report PRs to selectively accept, reject, or modify changeset entries. Triggered by /curate comment on knowledge/batch-* PRs. Reads PR comments, interprets natural language feedback, updates the changeset file, regenerates report, and commits.
Analyzes an Evidence Bundle and extracts knowledge candidates — the core LLM extraction step of the distillation pipeline. Stage B step 2. Transforms raw evidence into structured vault entry candidates by identifying confirmed team decisions, anti-patterns from incidents, and established conventions.
Queries team-verified knowledge from the Knowledge Vault. A UserPromptSubmit hook reminds you when active vault entries exist — use this skill to query and interpret them before planning code changes.
Executes bash commands
Hook triggers when Bash tool is used
Own this plugin?
Verify ownership to unlock analytics, metadata editing, and a verified badge. GitHub access is read-only (username + org membership).
Sign in to claimOwn this plugin?
Verify ownership to unlock analytics, metadata editing, and a verified badge. GitHub access is read-only (username + org membership).
Sign in to claimBased on adoption, maintenance, documentation, and repository signals. Not a security audit or endorsement.
A system that delivers only verified knowledge to AI coding agents.
When AI coding agents receive too much unfiltered context, they don't get smarter — they get noisier. Unverified hypotheses are treated as facts, revoked decisions persist as instructions, and irrelevant information buries the critical rules. The result: hallucinations, silent mistakes, and costly rework.
Knowledge Distillery takes a different path: collect broadly, distill rigorously, deliver minimally.
A 3-layer architecture with a convention-based air gap separates raw information from verified knowledge:
knowledge-gate CLIThe agent never touches raw data. It only sees what has survived distillation.
Executable skills in plugins/knowledge-distillery/skills/ are the source of truth for pipeline behavior.
| Document | Description |
|---|---|
| Design Philosophy | Why this architecture exists — the rationale behind information control for AI agents |
| Implementation Design | How it works — pipeline design, vault schema, runtime policies, deployment as a Claude Code Plugin |
| Adoption Guide | How to install the plugin, initialize a repository, and start using the vault |
| Tool Evaluation | Adopted and rejected tools with rationale |
The knowledge-gate CLI is the sole access path to the knowledge vault (.knowledge/vault.db). Set KNOWLEDGE_VAULT_PATH to override the default vault location.
Agent Runtime — query rules before modifying code:
| Command | Purpose |
|---|---|
query-paths [--ids-only] <filepath> | Resolve file path to domains → return a lightweight summary index, or IDs only |
query-domain [--ids-only] <domain> | Query rules by domain name → return a lightweight summary index, or IDs only |
search [--ids-only] <keyword> | FTS5 full-text keyword search → return a lightweight summary index, or IDs only |
get <id> | Retrieve full entry details (including body) |
get-many <id>... | Retrieve full details for multiple entries in one call |
list [--ids-only] | Summary list of all active entries, or IDs only |
Domain — explore and manage the domain registry:
| Command | Purpose |
|---|---|
domain-info <domain> | Domain details (description, patterns, entry count) |
domain-list [--status X] [--ids-only] | List domains or emit a lightweight domain ID index |
domain-resolve-path <filepath> | Reverse-lookup file path to domains |
domain-add, domain-merge, domain-split, domain-deprecate | Registry lifecycle |
domain-paths-set, domain-paths-add, domain-paths-remove | Path pattern management |
domain-report | Domain health diagnosis |
Loading — add entries to the vault:
| Command | Purpose |
|---|---|
add --type <type> --title ... --claim ... --body ... --domain ... [flags] | Add entry with validation |
_pipeline-insert | Pipeline-only bulk INSERT (JSON stdin) |
Utility — setup and maintenance:
| Command | Purpose |
|---|---|
init-db [path] | Create a new vault from bundled schema |
migrate | Apply schema migrations |
curate | Interactive curation queue resolution |
Examples:
# Query rules before editing a file
knowledge-gate query-paths src/api/auth/login.ts
# Query a domain with an ultra-light ID-only index
knowledge-gate query-domain --ids-only payment
# Fetch multiple full entries after a lightweight discovery step
knowledge-gate get-many global-convention service-rule
# Check which domains a file belongs to
knowledge-gate domain-resolve-path src/services/payment.rb
# Preload a lightweight domain ID index for navigation
knowledge-gate domain-list --ids-only
# Search for rules by keyword
knowledge-gate search "callback"
Run knowledge-gate help for full usage details.
herb-language-server support for .erb and .herb files
7 skills (1 gateway + 6 specialists) for building Rails frontend with Hotwire — Turbo Drive, Turbo Frames, Turbo Streams, Stimulus, view transitions, forms, media, and native bridge.
Autonomous product development loop — AI agent team that cycles through spec, implement, validate, and self-correct
Backlog.md task management workflow — setup, operating conventions, and guided task authoring.
Backlog.md task management workflow — setup, operating conventions, and guided task authoring.
npx claudepluginhub ether-moon/knowledge-distillery --plugin knowledge-distilleryAI-powered knowledge base management - Capture conversation learnings, maintain topic-specific KB files, Obsidian-compatible knowledge graph, structured dynamic context loading, and institutional knowledge in CLAUDE.md
Shared knowledge commons for AI agents; find, share, and confirm collective knowledge to stop rediscovering the same failures.
ARIA — Applied Reasoning and Insight Architecture. Persistent human-governed knowledge for Claude Code via a five-phase lifecycle: capture → govern → promote → apply → refresh. Stages session insights/decisions/feedback into review backlogs, promotes approved items to a tag-indexed markdown base, and applies them via /context, /rules, /codemap, /stitch, /distill, /prospect (forward-looking pre-mortems on plans before execution), /retrospect (commit/range/PR/release/deployment retrospectives with per-fix validation), and Rule 22 change-decision hooks on every Edit/Write. Both /prospect and /retrospect run an Evidence-Sourcing Pass that autonomously sources accessible evidence (codebase reads, public docs, MCP queries) and surfaces user-input asks for anything that requires judgment — converting unsupported assumptions to validated/falsified before the report finalizes. Reports persist to ~/knowledge/logs/{prospect,retrospect}/ with structured frontmatter and become discoverable via /context. Ships 34 working rules + a 7-step framework, audit cadences, ideas-lifecycle routing, drift detection, an optional shared-knowledge tier for team promotions, and structural-signal surfacing on critical paths. IMPORTANT: All skills except /setup require ~/.claude/aria-knowledge.local.md to exist. If missing when any skill is invoked, stop and tell the user to run /setup.
Knowledge base skills for Claude Code — capture, search, and synthesize project knowledge
Query Intent Solutions' governed knowledge brain from Claude Code — qmd://-cited answers, per-user audit, deterministic governance. Compile, then govern.
Automatic context engineering — observes your coding sessions and generates rules, suggestions, skills, and hooks so Claude gets smarter on your codebase over time