Recently updated open-source projects

This conversational skill guides structured self-reflection with one question per turn, plain pattern naming, concrete metaphors, and selectable personas such as Socratic, coach, mirror, or devil. It can use CBT, ACT, IFS, narrative, somatic, and other lenses, but is not therapy, diagnosis, or crisis support.

content-orchestrator routes a content request through research when freshness matters, WeChat article writing, humanization, and a final quality gate. It does not write the article itself; it coordinates the specialist skills and returns only the finished text, with a fallback when one child skill is unavailable. The intended result is source-supported, platform-aware copy rather than an unreviewed first draft.

debrief is a conversational learning ritual, not a document generator. It narrows the session to a few ideas, asks you to reconstruct each one in your own words one question at a time, reveals and corrects gaps only after the attempt, and distills a transferable principle with an optional durable note. It deliberately stays out of mechanical tasks, mid-task flow, and ordinary status recaps.

Feed interview-retro a recollection, transcript, or recording link after an interview. It produces either a learning-oriented question breakdown or a personal self-assessment with skill gaps and next actions, choosing the format from context. The default Feishu output needs an authenticated lark-cli setup, while local Markdown works without it.

Use the Idea Pyramid method to move from a business or social objective through audience perception, insight, strategy, creative direction, and execution. The Claude Code skill supports direct planning, clarifying questions, quick brainstorming, and a Feynman-style co-creation mode that challenges the user's initial thinking before producing a fuller plan.

Give the skill a character image and motion description to generate frames with Gemini, remove the background, align and time the sequence, and export looping SVG, WebP, and GIF files. A video path can reuse existing motion without a Gemini API key. Python and ImageMagick are required; interpolation and video input use ffmpeg, while ML background removal also needs onnxruntime, NumPy, and Pillow.

Regenerative Business Design guides founders through a Socratic dialogue, a written blueprint, or a diagnostic for an existing venture. It connects the Flourishing Business Canvas with regenerative principles, steward ownership, multi-stakeholder structures, nature and future-generation governance, and financing choices beyond standard venture capital. Recommendations include an explicit downside ledger, such as a smaller capital pool or slower raise, and the project states that local counsel is still required for legal and financial implementation.

This skill creates a clinical baseline Table 1 from XLSX or CSV data using design-aware rules: baseline p-values are hidden for RCTs by default, while observational studies emphasize SMD and multiple-testing warnings. It produces self-contained HTML plus Word and LaTeX files, masks identifier columns, and excludes survival, regression, and mixed-effects analyses.

kleinanzeigen-reader parses a kleinanzeigen.de URL into structured listing details and highlights signals such as commercial sellers, declared damage, missing inspection dates, unusually high mileage, or duplicate ads. It can compare listings, export JSON reports, and download listing images from either the CLI or a Claude skill. The Python 3.8+ package has no third-party dependencies.

Burnout Guard estimates burnout risk from self-report plus behavioral signals such as prompt heartbeats, late-night work, continuous blocks, and active-day streaks. Six levels range from quiet monitoring to single-task mode and a platform-level lockout with a cooldown, and Claude Code hooks can enforce the decision before a prompt reaches the model. State stays local, optional sleep and calendar inputs affect strain, and the project explicitly is not a medical device.

Use this framework as an agent skill or a manual checklist when reviewing data accounting, asset-backed securitization, property registration, trusted data spaces, and exchange-based transactions. It structures the work around regulatory comparisons, contradiction checks, a risk matrix, and a remediation list. The material is for reference and training rather than legal advice, so current rules and case-specific conclusions still need qualified review.

seocompass-skill brings SEO, GEO, and AEO workflows into Claude. It can find quick wins from Google Search Console, compare whether AI search engines recommend you or competitors, find backlink prospects and draft outreach, and run a hosted URL audit with a free SEOcompass key. Bring your own keys or use the hosted option; this DIY skill does not provide the app’s ongoing tracking and automation.

kw-asset-skill combines a Claude Code or Antigravity skill with a manual playbook for creating interactive HTML5 learning assets for the Knowledgeworker LMS. It uses Vite and vanilla JavaScript, includes portable templates and reference material, and can scaffold an asset project from a setup command with a selected type. The result is intended to run inside Knowledgeworker, so authors need access to that LMS and its embedding interface rather than a generic web host.

multi-agent-research routes a question through clarification, methodology and ethics checks, web, document, and academic scouts, source verification, synthesis, gap detection, adversarial review, and report writing. It rates or drops weak sources, preserves contradictions and uncovered areas, and returns a structured report in the language of the question. Seven phases and three quality gates make the workflow explicit, but the quality of the answer still depends on the available sources.

Indian Stock Analyst accepts portfolio exports from Zerodha, Groww, and other brokers, then uses live web research to summarize company news, regulatory signals, and retail sentiment. It produces Buy, Hold, or Avoid views for NSE and BSE holdings and flags sector concentration or structural risks. The workflow is intended for informational and educational use and is not a substitute for advice from a SEBI-registered professional.

roast-my-code analyzes a project across security, architecture, complexity, testing, type safety, performance, and other categories, then presents findings in a configurable roast style. It runs as a pure Claude Code skill without another API key. Its checks are strongest for JavaScript and TypeScript, with fewer ecosystem-specific rules elsewhere.

deep-think applies a staged reasoning loop to complex business, strategy, and decision questions: form hypotheses, examine counterarguments, test assumptions, self-evaluate, and rethink. It scores the response across coverage, logic, support, balance, specificity, and usefulness, adjusts effort to the question, and can wait for approval of the thinking frame. It relies mainly on supplied information rather than web research.

geng-skills combines deterministic statistics, duplication checks, and image inspection to flag suspicious data or edited figures in LaTeX or Word manuscripts before submission. The flagged material is then interpreted by an AI workflow, and the project exports a PDF with risk levels, image-forensics notes, an overall conclusion, and a disclaimer. It is a self-check aid, not a substitute for examining the original data and methods.