Recently updated open-source projects

cognitive-learning-canvas is a Chinese-language prompt framework for Claude, ChatGPT, Gemini, and similar assistants. It organizes difficult courses, papers, documents, or video transcripts into seven Obsidian Canvas layers: schema anchor, concrete example, terms, mechanism chain, self-test, deeper understanding, and a one-line close. A core metaphor and four-question test support pre-reading and concept mapping; it is not a replacement for the source or a step-by-step tutorial.

claude-credit-audit scans CI workflows and scripts for headless Claude calls, Claude Code Actions, and Agent SDK usage, then estimates monthly cost, per-call ranges, and the day a plan could run out of credit. It parses workflow triggers, model identifiers, and plan limits, and can calibrate token tiers from local session history. The project notes that the referenced billing change is currently paused, so the forecast describes a possible future impact rather than a live charge.

respec-editor helps update W3C ReSpec HTML specifications with correctly nested sections, definitions, normative statements, cross-references, and stable identifiers. It leaves generated tables of contents, bibliographies, indexes, and section numbers alone, and avoids changing metadata or normative status unless explicitly permitted. The skill prepares a branch, diff, and pull-request draft, then waits for confirmation before remote Git actions. It is scoped to ReSpec documents, not generic HTML, Bikeshed, or Markdown.

polymarket-data turns Polymarket’s public endpoints into research-ready JSONL for market metadata, outcome prices, and deduplicated trade history, with a calibration command for comparing price buckets to realized results. It handles the browser User-Agent requirement, the two-sided rows emitted for one fill, JSON-encoded market fields, and last-trade-per-minute history. The implementation uses only Python’s standard library and needs no API key or login. It is intended for public, unauthenticated data access.

Better OpenCode MCP exposes the OpenCode CLI to Claude Code and other MCP clients, enabling selectable models and non-blocking long-running tasks. Calls return a task ID that can be monitored or supplied with additional input, while process pooling, timeouts, and persisted task state help manage concurrency and recovery. It requires Node.js and a working OpenCode installation.

OpenCode Intern Skill lets Claude Code hand focused work such as refactoring or test writing to an OpenCode agent through the Agent Communication Protocol. Both tools can share project instructions and skills from the same workspace, while the intern is configured to follow the requested scope and report blockers instead of making broad decisions. Claude Code, OpenCode, and Python 3.8 or later must be installed.

Scholar Sidekick accepts identifiers such as DOI, PMID, ISBN, ISSN, arXiv IDs, and WHO IRIS URLs, then formats citations in more than 10,000 CSL styles. Its skills can export BibTeX, RIS, EndNote, CSL-JSON, or CSV and check retractions, open access, and fabricated metadata; it resolves known identifiers rather than searching for papers by topic.

grill-with-docs-html adds an interactive HTML form output to the `grill-with-docs` Socratic interview. It presents facts with file and line references, decision cards, trade-offs, WHY and side-effect panels, and validated Before/After Mermaid diagrams, then copies structured Markdown to the clipboard. The base `grill-with-docs` skill, Node.js, and Mermaid CLI are required.

Run this skill after a development phase to assign two to seven fresh subagents across specification, architecture, code quality, security, and performance. It selects an S, M, or L workflow from diff size and sensitive areas; larger reviews can add an adversarial skeptic and a fresh regression sweep after fixes. Changes touching authentication, PII, payments, uploads, or raw SQL are bumped to at least medium depth. It is designed for phase boundaries, not CI, PR review, or a replacement for human security review.

seo-blog-pipeline rotates through configured content silos, finds uncovered low-competition questions from SERP data, and structures answer-first posts for WordPress. It deduplicates topics in a local ledger, checks planned internal links, uploads a featured image with alt text, and creates a draft or published post through the WordPress REST API. A WordPress application password and a SERP provider such as Apify are required; optional video and image providers can enrich the result.

This skill helps Claude handle ServiceNow development and architecture requests with platform-specific patterns instead of generic JavaScript advice. It separates domain references from coding standards for areas such as GlideRecord, Flow Designer, security, and performance, then produces ServiceNow-specific code, workflow or architecture guidance with relevant role, plugin, testing, and deployment assumptions.

business-analysis-generator researches a company and its market, uses the findings to shape an adaptive interview, and turns the result into four to ten reusable analysis prompts. Each prompt specifies the business connectors, grounding choices, useful response, failure modes, and industry adaptations, making the library suitable for consulting audits and implementation planning. The company analysis can be refreshed occasionally, while the prompts can run on a recurring cadence against live business data.

connect-gemini-skill makes a locally authenticated Gemini CLI available inside Claude Code through /gemini, /gemini-review, and a dispatchable Gemini subagent. It supports independent opinions, Google-grounded research, very large context analysis, and diff review without switching terminals. The installer checks or installs the CLI, verifies authentication, and places the connector files. The coding-focused CLI can analyze images but does not generate Nano Banana images, and it targets macOS or Linux.

Quick Symlink adds common link operations to the macOS Finder context menu, avoiding repeated terminal commands. You can create links for selected files or folders, copy a source and paste its link into another directory, or move an item while replacing its original location with a symlink. After installing the app, the relevant Finder extensions must be enabled in system settings, and users can choose whether newly created symbolic links use relative or absolute paths.

Create a JSON storyboard with objectives, a narrative arc, and one message per slide, then render it deterministically to a static PowerPoint file with PptxGenJS. Business and education modes apply different storytelling or learning structures, while the QA step checks titles, slide density, citations, transitions, and objective coverage. Node.js 18 or newer is required.

HBR-Style Visualization provides Claude with design rules, prompt templates, and a quality checklist for flywheels, maturity models, annotated charts, and 2×2 strategy matrices. It centers each graphic on a clear argument and uses restrained color, serif typography, and clean layouts for executive presentations, research figures, and business reports.

Humanizer-JA rewrites Japanese produced by ChatGPT, Gemini, or Claude so it reads more naturally. The Claude Code skill checks 30 patterns across six categories, including repetitive rhythm, manual-like structure, vague claims, leftover formatting, and awkward metaphors, with modes for everyday publishing, business, academic, medical, and diagnostic use.

Akanuke detects recurring LLM writing habits in English and Japanese, including inflated claims, promotional tone, excessive hedging, rigid formatting, monotonous rhythm, and clichéd metaphors, then rewrites the text in general, formal, academic, or medical modes. An optional report explains the detected patterns; the result should still be checked for preserved meaning and facts.