Shape agent capabilities into your workflow.

consumer-dispute-assistant organizes messy refund, subscription, and billing problems into a usable case summary, merchant request, payment-dispute statement, complaint draft, and follow-up log. It helps users preserve a factual timeline and escalate gradually instead of producing only a generic complaint letter. The skill is not legal advice, does not invent evidence, and leaves every external submission under the user's control.

Curator triages session context before a compressor decides what to shorten or discard. It scores information by recoverability, constraint criticality, and dependency risk, then routes it as pinned, structurally preserved, summarizable, or discardable. Use it to keep prohibitions, exact configuration values, and decision rationale visible across long agent sessions while continuing to use a separate compression tool for lower-risk content.

Build a structured voice profile from your own writing samples, including tone, sentence patterns, formatting habits, and phrases to avoid. ResoNote then scouts topical angles and creates separate drafts for platforms such as X, LinkedIn, Bluesky, Threads, Instagram, TikTok, and Mastodon. Posts remain in a human review queue before publishing, and engagement data can refine the profile; direct publishing may require third-party APIs.

Give The Board a technical, product, venture, launch, or risk decision, and it selects a panel of contrasting advisory archetypes to review feasibility, economics, user value, compliance, and failure modes. Their independent votes are aggregated into PROCEED, CONCERN, or BLOCK under explicit rules, with reasons and conditions for reconsideration. The panel uses composite roles rather than imitating named real people.

Project Foundations adds language-specific setup guidance to Codex and Claude project work without replacing the existing superpowers workflow. Its references cover project structure, naming and style, tooling, starter configuration, TDD, CI checks, quality gates, and common review points across many languages. An installer registers the plugin with Codex and places the marketplace entry, while the same package can be installed for Claude. It is useful when new projects need consistent starting conventions.

secauditor examines a codebase without editing source or running exploits, scoring up to eleven security dimensions against OWASP, CWE, SLSA, and relevant NIST guidance. It writes a self-contained secaudit.md with exact file locations, source-to-sink paths, preconditions, impact, a specific fix, and a way to verify the change. Conditional dimensions are omitted when a project has no matching surface, and the audit looks for controls that exist only on paper. It runs in Claude Code or Codex.

Describe an architecture task in Claude Code and the skill selects an appropriate combination of C4, ArchiMate, TOGAF, or arc42 with ADRs. It can produce diagrams, documentation, decision records, graded reviews, and enterprise models while reusing stable element IDs across artifacts. A companion linter checks cross-document consistency, including undefined IDs, broken ADR links, and orphaned diagram sources.

This Bash wrapper reduces a Google Cloud GPU workflow to creating a VM, pushing code, launching detached jobs in tmux, streaming logs, pulling results, and stopping or deleting the instance. It works through Claude Code or Codex as well as a standalone CLI. You still need a configured gcloud login, a project with GPU quota, and attention to shutting down running machines because billing continues while they are active.

shap-figures provides a reusable Python library for publication-oriented SHAP, GAM, and GIS visualizations. It includes coolwarm beeswarms, feature-importance charts, GAM dependence plots with confidence intervals, predicted-versus-actual views, spatial SHAP maps, trends, ablation grids, and labelled multi-panel composites. The package uses a consistent visual style and can be called directly from Python. NumPy, pandas, Matplotlib, and SciPy are required; mapping and image-stacking features add optional dependencies.

VisionCraft lets Claude Code use Agnes AI to create images and videos from text or reference images. It supports multi-image and keyframe workflows, handles polling and timeout recovery, and validates JSON, but requires Windows PowerShell 7+, a valid Agnes API key, and the supported session environment.

solana-fundraise-skill helps Solana projects choose suitable funding programs, write credible applications, and plan delivery after funding. It covers positioning, program fit, proposals, social proof, outreach, rejection handling, and grant stacking across Solana Foundation, Superteam Earn, and Colosseum.

This Claude Code plugin helps developers set up Lore, run everyday version-control commands, understand its content-addressed model, and work with the Python, JavaScript, C#, and Go SDKs. It also maps common failures to practical fixes and advises checking the installed CLI help because Lore is still pre-1.0. The project is community maintained and is not endorsed by Epic Games.

Run the skill in an existing codebase to create or audit root and per-directory CLAUDE.md files from the repository's actual architecture, dependencies, and conventions. It also generates four specialized agents for analysis, implementation, testing, and documentation, plus a guide for handing tasks through them in order. Because initialization writes files autonomously and the project is still early, users should preserve a rollback point and review the resulting diff.

The two skills combine a conventional on-page SEO pass with a GEO pass focused on making useful passages easier for AI answer engines to retrieve and cite. They can audit a page, improve intent and metadata, restructure copy into extractable answer blocks, and examine competitor citation gaps. A dedicated MENA and Arabic playbook covers multilingual queries and entity consistency, while reminding users to recheck fast-changing citation behavior and statistics.

Anti-Apophenia makes an agent treat an appealing pattern as a hypothesis rather than a finding. It requires a claim, evidence, null model, alternatives, prediction, and disconfirmation loop, while penalizing post-hoc fitting, repeated attempts, cherry-picking, and vague resemblance. Evidence grades and explicit verdicts separate strong, weak, rejected, and unknown results, and tempting branches stay parked instead of steering the investigation. Speculative exploration is allowed only in a clearly marked sandbox.

hekouwang-stock-data-reader-skill accepts an A-share code, pulls public market and company data through AkShare, calculates a fixed set of indicators, and produces publishable chart cards plus long-form and headline versions. Its built-in boundary is deliberately non-predictive: it does not recommend stocks, forecast prices, or give trading points, and adds a risk notice to the output. The data source needs no API key, but the result remains an informational retrospective rather than investment advice.

github-auto-backup interviews you about repository settings when creating a project, then manages local Git and GitHub backups with conventional commits. It can commit, push, pull, and create branches, but pauses before public pushes, force-pushes, opening or merging pull requests, deleting branches, or changing repository settings. It never commits secrets or silently makes a private repository public. Local-only mode works without GitHub; GitHub operations need an authenticated gh CLI.

Living Docs treats documentation as a governed part of the codebase, organizing constitutions, PRDs, ADRs, BDRs, issues, research notes, and Mermaid diagrams around five invariants. Facts have one home, every document is indexed, old decisions are superseded rather than rewritten, and structural changes carry their documentation in the same change. A deterministic linter checks the contract across Claude Code, Cursor, Copilot, OpenCode, Codex, and Pi. It governs organization, not a project’s technology choice.