adapter
Route to, register, inspect, validate, and run explicitly trusted local private adapters…
Capture, research, shape, catalog, and explicitly promote Ideas into Projects through natural language.
> /plugin marketplace add nvk/llm-wiki > /plugin install wiki@llm-wiki
How it fires
How this command gets triggered: by you, by Claude, or both.
/ideaContext preview
What this command does when you run it.
Capture, research, shape, catalog, and explicitly promote Ideas into Projects through natural language.
description: "Capture, research, shape, catalog, and explicitly promote Ideas into Projects through natural language." argument-hint: "new [<slug>] \"<seed>\" | list [filters] | show|develop|shape|promote|archive <slug> [--wiki <name>] [--local]" allowed-tools: Read, Write, Edit, Glob, Grep, WebFetch, WebSearch, Bash(ls:*), Bash(wc:*), Bash(date:*), Bash(mkdir:*)
Manage Ideas between compiled Concepts and committed Projects. Read `skills/wiki-manager/references/ideas.md`; load inventory, research, or Project references only for the operation that needs them.
Resolve HUB from `$HOME/.config/llm-wiki/config.json` (`hub_path` first), then `--local`, `--wiki`, CWD `.wiki/`, or HUB. Read indexes first. Writes require one active topic and never go at hub root.
Natural language is primary. Infer `new`, `list`, `show`, `develop`, `shape`, `promote`, or `archive`; resolve phrases such as "the podcast thing" from the conversation, titles, aliases, and Idea indexes. Derive safe slugs.
Check Idea titles/aliases, relevant Concepts/inventory, and Project slugs for duplicates. Link related Ideas; never merge automatically.
1. Lazily create `inventory/ideas/` and its `_index.md`. 2. Write `inventory/ideas/<slug>.md` with required inventory fields, `kind: idea`, `status: proposed`, default `priority: p2`, `next_action`, and related wiki/source links when known. 3. Preserve the exact seed under `## Original Seed`. Add concise sections for opportunity, target, evidence, assumptions/counterevidence, alternatives, current shape, scope/no-gos, risks, questions, success criteria, intended deliverable, and decision notes. 4. Rebuild indexes and log `idea | captured <slug>`.
Do not duplicate the seed as a raw note or create a Project.
Read the Idea index first; hub-wide lists read only active topics' Idea indexes. Filter by status, maturity, priority, topic, tags/aliases, owner, next action, or Project. Default to a short list.
Derive maturity: captured when the record exists; researching from linked raw research; shaped from `approved_at` plus an Approved Decision Snapshot; project from a linked non-empty `WHY.md`; delivered from a linked outcome/deliverable. Never add a synchronized `stage` field.
Read the Idea, linked Concepts, and indexes; research only gaps. Evidence goes in `raw/`, reusable synthesis in `wiki/`, and the Idea distinguishes evidence, assumptions, contradictions, and unknowns. Update status/date/next action, indexes, and log. Research never implies approval or Project creation.
Challenge the premise. Write minimal, ideal, and optionally lateral versions; define target, appetite, scope, no-gos, risks, rabbit holes, success criteria, and deliverable. Without a user choice, set the next action and stop for a concise decision. With explicit approval of a named version, add `approved_at: YYYY-MM-DD` and `## Approved Decision Snapshot` containing the chosen rationale, scope, no-gos, success criteria, and deliverable. Update indexes and log (`idea | shaped <slug>`).
Require explicit approval in the current request or both `approved_at` and an Approved Decision Snapshot. Otherwise ask for approval. Promotion is idempotent:
1. Show an already-linked valid Project; do not duplicate it. 2. Refuse active/archived Project slug collisions. 3. Create `output/projects/<slug>/WHY.md` from the approved rationale with a backlink, and freeze the snapshot as `BRIEF.md` with a backlink. 4. Add `project: output/projects/<slug>` and `promoted: YYYY-MM-DD` to the Idea, update its next action/indexes, and log `idea | promoted <slug> -> output/projects/<slug>`.
The Idea remains catalog/lineage. Plans, implementation, artifacts, outcomes, and changing delivery truth belong to the Project.
`archive` sets Idea `status: archived`, date, decision notes, indexes, and log; it never archives a linked Project. If fuzzy wording could mean Idea, Project, or both, ask before changing state.
Report path, topic, status, maturity, links, Project, and next action.
LLM-compiled knowledge bases for any AI agent. Parallel multi-agent research, thesis-driven investigation, source ingestion, wiki compilation, querying, and artifact generation.
Route to, register, inspect, validate, and run explicitly trusted local private adapters…
Archive or restore whole topic wikis so old interests stay preserved but out of default…
Assess a local repo against the active wiki's research body and the broader market. Gap…
Truth-seeking umbrella audit for llm-wiki. Combines active wiki maintenance, output drift…
Export, refresh, verify, or import a comprehensive cross-topic Project Knowledge Checkpoint…