Browse the directory

Run AI inference and serverless functions on Cloudflare Workers AI: call hosted models like Llama, Whisper, and Stable Diffusion through the Workers AI binding, deploy with wrangler, and use D1/R2/KV storage plus the free daily Neuron allocation.

Community slash command runbook to refresh stale project documentation after code changes: use git history to find affected docs, compare README commands to package scripts, and flag broken internal links before opening a docs PR.

Community slash command runbook for OpenAPI contract review: lint base and head spec files with the documented Spectral CLI, compare rule failures, and classify release impact before merging API changes.

Community slash command runbook for adding minimal automated tests around a changed module: inspect the git diff, mirror repository test conventions, and draft focused unit or integration tests using Anthropic develop-tests guidance.

Slash command that reviews a pull request diff for security regressions: authentication and authorization gaps, injection surfaces, secret exposure, unsafe deserialization, and dependency risk introduced by the change.

Local-first MCP server for web scraping, crawling, URL mapping, batch extraction, structured extraction, summarization, diffing, brand extraction, and optional hosted API fallback for bot-protected pages.

Source-backed agent for reviewing local-first Jupyter notebook workflows that use Jupytext text notebooks, paired notebooks, command-line sync, and version-control friendly reproducibility checks.

Source-backed rules for registry repositories that must keep contributor PRs focused on source files while generated indexes, feeds, downloads, search data, README output, and previews are rebuilt by trusted automation.

Apache-2.0 library for pretrained diffusion model pipelines, schedulers, adapters, optimization, and training workflows for image, video, and audio generation in PyTorch.

PostToolUse hook that flags two review hazards as Claude writes files: edits to generated or vendored artifacts (lockfiles, minified bundles, source maps, dist/build output, snapshots, protobuf output) and oversized diffs that change more lines than a configurable threshold versus the committed version.

PostToolUse hook that catches unintended UI changes by pixel-diffing a just-saved screenshot against its baseline with odiff.

Read-only Claude Code PostToolUse hook that runs shfmt diff checks after Claude writes or edits shell scripts, surfacing POSIX shell, Bash, Zsh, and mksh formatting drift without rewriting files or executing scripts.

Validates translation files for missing keys and ensures consistency across different language files.

Expert reg-suit review skill for evaluating rendered UI image baselines, thresholds, snapshot storage, report artifacts, and visual QA release readiness.

Source-backed agent that operates Claude Code inside GitLab CI pipelines — triaging pipeline failures, generating MR descriptions, running automated code review on diffs, and reporting findings back to merge requests via the GitLab API through a narrow write proxy and explicit prompt-injection boundaries.

MCP server for parsing Korean HWP, HWPX, HWPML, PDF, XLSX, and DOCX files into Markdown, with tools for metadata extraction, page ranges, tables, document diffs, form extraction, and HWPX form filling.

A practical walkthrough of Claude Code checkpointing: how automatic checkpoints capture code before each edit, how to rewind and summarize with /rewind, the difference between restore and summarize, and the limitations you must know before a risky refactor.

Local MCP server that gives AI assistants terminal control, process management, filesystem operations, search, diff editing, and document handling for desktop automation and development workflows.

A practical walkthrough of least-privilege permission design in the Claude Agent SDK. Learn the evaluation order (hooks → deny → mode → allow → canUseTool), how allowedTools and disallowedTools differ, the six permission modes, and how to lock an agent to a fixed tool surface with dontAsk.

A practical walkthrough of managing Claude Code sessions: resuming with --continue, --resume, and --from-pr, naming sessions, using the session picker, branching to try a different approach, and where transcripts are stored.

A source-backed review workflow for pull requests that include AI-generated code. Treat generated diffs as untrusted implementation work, verify behavior in CI, inspect security-sensitive paths first, and merge only after a reviewer-owned checklist passes.

Source-backed rules for reviewing TypeScript API client compatibility before merge, with exported type-surface diffs, inferred router inputs and outputs, runtime validator alignment, downstream compile checks, and privacy-safe evidence.

Apache-2.0 library for parameter-efficient fine-tuning of large pretrained models with adapters, LoRA, prompt tuning, Transformers, Diffusers, and Accelerate.

Alerts when files exceed size thresholds that could impact performance. This PostToolUse hook provides comprehensive file size monitoring when files are created or modified, automatically detecting files that exceed recommended size thresholds for different file types and providing optimization suggestions.

Install and use the official Claude Code security-guidance plugin before merge: per-edit pattern warnings, end-of-turn and commit git-diff review, and team rollout for command injection, XSS, eval, and dangerous file edits.