Core workflow for an Open Knowledge Format knowledge base. Initialise a new bundle or adopt an existing folder of markdown, compile raw sources into a self-contained wiki with tracked provenance, check its health, and record human sign-off on an article. Searching the wiki lives in the kb-query plugin. Requires the okf-kb Python package for the kb-* tools.
Plugin kb from marketplace okf-kb in carelvniekerk/okf-kb. Commands read from the repo's manifest — 12 days ago.
This plugin runs code on your machine — its hooks fire automatically on events like tool use. The exact commands are listed below; read them before installing.
adoptTake over an existing folder of markdown as an Open Knowledge Format knowledge base. Surveys what is there, separates sources from articles, writes okf.toml describing the layout it found, backfills OKF frontmatter with provenance recovered from git history, and gets kb-health green without rewritin
compileRun the knowledge base compilation pipeline. Transcribes handwritten notes, converts PDFs, Office documents and e-books, detects new, modified and deleted sources since the last compile, integrates each into the wiki with a strategy chosen by source type, regenerates the indexes, runs the health che
healthRun a full wiki health check. Runs the automated link, image and provenance checks, then reviews the wiki for undefined concepts, contradictions between articles and stale content, and reports the findings by severity without fixing anything. Use when the user says "health check", "lint the wiki" or
initScaffold a new Open Knowledge Format knowledge base in an empty directory. Checks the kb-* tooling against the enabled plugins, asks what the bundle is for, writes okf.toml, CLAUDE.md, .claude/settings.json, the directory skeleton, .gitignore and optional pre-commit and VS Code files, then generates
verifyRecord a human sign-off on a wiki article. Reads the article back against every source it cites, reports drift and unsupported claims, and only after the user's explicit confirmation appends an OKF verified entry with their human id. Use when the user says "verify", "sign off on" or "review this art
source-reviewerReads one wiki article back against every source it cites and reports drift, misattribution, wrong detail and stale claims, with quoted evidence. Use when /kb:verify needs the article-against-sources comparison, or when the user asks for a fresh read of an article's claims against its raw sources.
uv
uv
Daily capture for an Open Knowledge Format knowledge base: start-of-day briefs that pull your calendars and inbox, meeting notes scaffolded from the Google or Microsoft 365 calendar event, and voice notes, all filed into raw/ as ordinary sources. Calendars and mail accounts are configured per bundle in the [capture] table of okf.toml, so nothing about your accounts is built into the skills.
3 skills
Source acquisition for an Open Knowledge Format knowledge base: fetch arXiv papers, convert PDFs, Word, PowerPoint, Excel, EPUB and Outlook files to markdown (PDFs with their figures), clip web pages verbatim, download referenced images locally, and transcribe handwritten notes with vision. Needs the okf-kb package's [ingest] extra, which brings pymupdf, requests, beautifulsoup4, markdownify and markitdown.
2 skills
Read-only access to Open Knowledge Format knowledge bases from any project. One skill, wiki, searches the knowledge bases in scope, follows their cross-links and backlinks, and reads the articles it needs, so an agent checks what you have already written down before answering from training data. Scope comes from okf.toml when you are inside a bundle, or from a project's .okf-kb.toml and ~/.config/okf-kb/config.toml when you are not. Nothing in this plugin can write to a knowledge base. Its tools come from the okf-kb MCP server (kb-mcp), which the plugin starts and the Claude desktop app can register too. Requires the okf-kb Python package.
1 skill · 1 MCP server
YouTube ingestion for an Open Knowledge Format knowledge base. Stages a video's metadata, captions, audio and frames, judges transcript quality and falls back to MLX Whisper, then writes a structured article with selected frames and keeps the verbatim transcript beside it. Needs the okf-kb package's [video] extra (yt-dlp, mlx-whisper) and ffmpeg on the system.
1 skill
Personalized AWS guidance built on patterns from 350,000+ startups. Get architecture, cost, and security recommendations matched to your stage, from day-one account setup to production-ready infrastructure. Migrate to AWS from Azure, GCP, or Heroku with resource discovery, cost estimates, and generated Terraform. Move OpenAI, Azure OpenAI, and Gemini code to Amazon Bedrock, and plan and prototype AI agents on AgentCore and other AWS runtimes. Includes AWS Activate credits eligibility, 60+ exclusive startup offers, and multi-account, multi-region support. Built by AWS Startup Solutions Architects.
13 skills · 1 command · 7 agents · 1 MCP server
Search across all of your company's tools in one place. Find anything across email, chat, documents, and wikis without switching between apps.
5 skills · 7 MCP servers
Personal knowledge brain for your coding agent — hybrid search, synthesis, and durable cross-session memory, plus a curated brain-first skill set.
80 skills · 1 MCP server · hooks
Build, launch, grade, and schedule Claude Managed Agents (CMA) in your own Anthropic account — a plugin re-implementation of Anthropic's launch-your-agent reference skill (Apache-2.0; independent, not a fork). Every session starts with a goal (./my-agent/goal.json, opt-in SessionStart hook, driven by /cs-goal); loop_compiler.py compiles it into a bounded grade->iterate loop (max_iterations 1..20), a recurring cron scheduled-deployment loop, or a single-pass interview->stage->launch workflow. 6 skills: orchestrator (context: fork goal router) + interview + stage-launch + grade-iterate + run-without-you + wrap-up. 18 stdlib-only deterministic scaffolders (no network/API calls; live launches emitted as BYOK curl that never prints the key), 4 agents, 8 commands, hooks, 5 references, 4 assets. Validators enforce CMA limits. Distinct from engineering/agent-harness (generic domain loop) and write-a-skill (authors Claude Code skills, not CMAs).
6 skills · 8 commands · 4 agents · hooks
Brainstorm, plan, debug, review, and compound learnings with AI agents
39 skills
Answer technical questions using daily.dev's knowledge base. Searches articles from the developer community and synthesizes answers with source links.
6 skills
This page updates itself from your repo. Add the badge to your README, or get it Featured to pin it above the organic results with a followed link to your site.