Help us improve
Share bugs, ideas, or general feedback.
Share bugs, ideas, or general feedback.
Share bugs, ideas, or general feedback.
By cosmicdreams
Passive knowledge capture pipeline — Karpathy-derived LLM Wiki with Monitor-driven multi-domain ingest. Feeds land in raw/, agents compile them into a cross-linked Obsidian wiki, memory evolves via confidence decay and graph-aware consolidation.
npx claudepluginhub cosmicdreams/claude-plugins --plugin ideas-funnelDomain-scoped ingestion worker. Spawned by the orchestrator per batch_complete or manual-ingest signal. Reads Raw/Inbox/<domain>/, runs the /ideas-funnel:ingest skill, writes domain-scoped wiki pages, updates the manifest, and emits wiki/concept-density-threshold signals when a concept crosses ≥3 unrelated sources.
Ephemeral per-signal coordinator for the ideas-funnel pipeline. Reads a single Monitor signal line, creates a short-lived team, spawns the right domain ingest-agents (one per affected domain, in parallel), conditionally spawns Refinery and scorer agents, collects subagent completions, writes an events JSONL line, deletes the team, and exits. Not resident between firings.
The single writer for all shared wiki layers (Concepts/, Entities/, Bridges/, Conflicts/, _meta/conflicts.md). Spawned by the orchestrator on concept-density-threshold or bridge-threshold-crossed signals. Consolidates multi-source concepts into synthesis pages, detects contradictions, writes bridge pages for cross-domain concepts. Serializes writes to prevent decoherence from parallel ingest-agents.
Processes unprocessed items in Raw/ and Raw/Inbox/<domain>/ — breaks each into Source, Concept, and Entity pages; updates index.md, log.md, and .manifest.json; emits wiki/concept-density-threshold signal when a concept crosses the ≥3-unrelated-sources bar. Extends the existing vault ingest skill with multi-domain dispatch and the v2 frontmatter schema (timeline, confidence, decay_class, state, provenance). Trigger phrases: "ingest", "process raw items", "process inbox", "/ideas-funnel:ingest".
One-time bootstrap for the ideas-funnel plugin. Creates the user config directory, writes a starter domain YAML if none exist, verifies the vault has the required scaffold (Raw/Inbox, Domains/, _meta/, CRITICAL_FACTS.md), appends the v2 schema extension to wiki-schema.md if not already present, and prints a next-steps checklist. Trigger phrases: "init ideas-funnel", "bootstrap the funnel", "/ideas-funnel:init".
Wiki-aware vault linter. Scans for structural violations, schema compliance (v1 + v2 fields), index.md sync, Raw/ staleness, orphan pages, and timeline overflow. Reports grouped by severity. Writes `^stale` flags and `backlink_density` to page frontmatter. Trigger phrases: "lint the vault", "vault health", "/ideas-funnel:lint", "check the vault".
Answer a question against the vault. Scans index.md for candidate pages, reads the relevant bodies, synthesizes a response with [[wikilink]] citations, and optionally files the answer back as a new Concept page if the answer is novel enough to deserve one. Trigger phrases: "query the vault", "/ideas-funnel:query", "what does the vault say about X".
Uses power tools
Uses Bash, Write, or Edit tools
Share bugs, ideas, or general feedback.
Own this plugin?
Verify ownership to unlock analytics, metadata editing, and a verified badge.
Sign in to claimOwn this plugin?
Verify ownership to unlock analytics, metadata editing, and a verified badge.
Sign in to claimBased on adoption, maintenance, documentation, and repository signals. Not a security audit or endorsement.
Upstash Context7 MCP server for up-to-date documentation lookup. Pull version-specific documentation and code examples directly from source repositories into your LLM context.
AI-powered wiki generator for code repositories. Generates comprehensive, Mermaid-rich documentation with dark-mode VitePress sites, onboarding guides, deep research, and source citations. Inspired by OpenDeepWiki and deepwiki-open.
Claude + Obsidian knowledge companion. Sets up a persistent, compounding wiki vault (Karpathy's LLM Wiki pattern). v1.7 "Compound Vault" + v1.8 methodology modes close 5 of 5 priority gaps from the May 2026 compass artifact. Ships: substrate alignment with kepano/obsidian-skills, default Obsidian CLI transport, hybrid retrieval (contextual prefix + BM25 + cosine rerank per Anthropic's Sept 2024 research), per-file advisory locking for multi-writer safety, pre-commit verifier agent, AND methodology modes (LYT / PARA / Zettelkasten / Generic) for first-class organizational support no other Claude+Obsidian competitor offers. v1.7.x audit closure: every BLOCKER + HIGH + MEDIUM + LOW finding from the v1.7.0 audit is CLOSED or DEFERRED-with-rationale. Optional DragonScale Memory extension (log folds, deterministic addresses, semantic tiling lint, boundary-first autoresearch).
Complete AI coding workflow system. Self-correcting memory + persistent FTS5-indexed research wikis + auto-research loop + multi-LLM council on a single SQLite store. 33 skills, 8 agents, 22 commands, 37 hook scripts across 24 events. Cross-agent via SkillKit.
Make your AI agent code with your project's architecture, rules, and decisions.
Team sprint execution infrastructure: agents, skills, hooks, protocols, and kanban pipeline.
Plugin, agent, and skill management tooling: bump-version, scaffold, new-agent, new-skill, optimize-agents, update-plugins, create-worktree.
Sprint retrospective infrastructure: agent interviews via SubagentStop hook, session analysis, action cards, and retrospective kanban board.
Pre-work ideation with a visual decision canvas. Generates divergent ideas, launches a local web UI for human annotation, then synthesizes decisions into a concrete plan.
Drupal-specific development skills and agents (DDEV, Analysis, Validation, Scaffolding).
A collection of Claude Code plugins covering team sprint orchestration, Drupal development, process engineering, passive knowledge capture, and meta-tooling for plugin authoring.
Several plugins require external CLI tools (Beads, Obsidian CLI, GitHub CLI, jira-cli, ddev, etc.).
→ See DEPENDENCIES.md for the full install guide.
# Clone the repo
git clone git@github.com:cosmicdreams/claude-plugins.git
cd claude-plugins/worktrees/main
# Install plugins at user scope
claude plugin install admin@local --scope user
claude plugin install sprint@local --scope user
claude plugin install retro@local --scope user
claude plugin install ideate@local --scope user
claude plugin install drupal-lab@local --scope user
claude plugin install lib@local --scope user
claude plugin install workflow@local --scope user
claude plugin install drover@local --scope user
claude plugin install research-lab@local --scope user
claude plugin install improve@local --scope user
claude plugin install ideas-funnel@local --scope user
After installing, initialize the sprint board in your project:
brew install beads
bd init --prefix sprint
adminMeta-tooling for developing and maintaining Claude Code plugins.
Skills: agent-team, bump-version, changelog, create-worktree, install, new-agent, new-skill, optimize-agents, scaffold, scaffold-silence, update-plugins
sprintTeam sprint execution: parallel agents, kanban pipeline, hooks, protocols.
Skills: asset-audit, board, kanban, plan, project-notes, run
retroEnd-of-sprint retrospectives: agent interviews via SubagentStop hook, action card management, session reports.
Skills: interviews, kanban, session, transcript
ideatePre-work ideation: brainstorm canvas, deep research, structured comparison, reality checks, diagrams, ADRs.
Skills: adr, brainstorm, compare, diagram, reality-check, research, understand
drupal-labDrupal development against DDEV: issue analysis, patch validation, contrib module scaffolding, performance profiling.
Skills: analyze-issue, browse-drupal-issues, config, ddev, finish-issue, issue-summary, module-dev-starter, perf-measure, process-lifecycle, validate-patch
libThin CLI-wrapper skills (data-layer only — no summarization). Slack, Jira, GitHub, TestRail, Obsidian vault, logs, and media utilities.
Skills: archive, csv-analysis, ddev, ffmpeg, github, hyperfine, image-optimize, jira, lighthouse, log-analyzer, pa11y, slack, testrail, vault-search, vault-store, wiki-query
workflowProcess automation built on top of lib: morning brief, deploy checklist, cross-source pulse, Obsidian maintenance, calendar/email helpers.
Skills: config, deploy-post, ecosystem-pulse, morning-brief, obsidian-lint, organize, personal-calendar, personal-email, pulse
droverAutomated Drupal error monitoring and self-healing pipeline. Watches logs, triages errors into curated Beads tickets, autonomously implements fixes in isolated git worktrees, notifies on ready-for-review.
Skills: add-project, backfill, baseline, board, dashboard, implement, reset-state, run, setup, triage, watch
research-labComposable research engagement pipeline: literary review via NotebookLM, multi-agent workshop swarms, cross-examination seminars, and iterative experimentation with ratchet-based optimization.
Skills: experiment, literary-review, run, seminar, workshop
improveProcess engineering methodology. Maps process topology, makes directed fixes, runs improvement experiments, accumulates lint rules. Domain-agnostic — each plugin can own its own :improve skill for domain-specific knowledge.
Skills: accessibility-scan, attach, experiment, fix, lint, perf-measure, self
ideas-funnelPassive knowledge capture pipeline — Karpathy-derived LLM Wiki with Monitor-driven multi-domain ingest. Feeds land in Raw/, agents compile them into a cross-linked Obsidian wiki, memory evolves via confidence decay and graph-aware consolidation.
Skills: ingest, init, lint, query
admin:changelog <plugin> # e.g. admin:changelog sprint
admin:changelog <plugin> --latest # most recent version only
Chris Weber