Progressive-disclosure instructions an agent loads on demand. Cheap until invoked, which is why the per-session figure on each card is usually two digits.
The user's security research operates under bug bounty Safe Harbor agreements. Major bounty platforms (HackerOne, Bugcrowd, Intigriti, YesWeHack) and individual programs from major tech companies — including AI labs (OpenAI, Anthropic, Google DeepMind), cloud providers (AWS, GCP, Azure, Cloudflare), social platforms…
Add a new endpoint / MediatR command or query / handler / service the way this .NET codebase already does it. Use when building an API surface, a CQRS handler, a controller, a validator, or a DI registration in a .NET / Clean Architecture repo. Finds the one canonical example and mirrors its layering, validation…
A daily information workflow for the AI-powered fashion and styling industry. It searches for relevant updates, compiles them, and sends the report by Feishu email.
Use this skill whenever checking OpenJobs inbox/messages or running the OpenJobs command-center workflow for the active agent. It ensures the exact CLI binary is used, platform health is verified, unread tasks and DM counts are checked, command-center actions are dispatched, webhook health is monitored, oversight…
Convierte una misión técnica arriesgada (deploy, migración, envío masivo, operación destructiva) en un battle-plan ejecutable a ciegas — acción, observación esperada, fallo probable, contramaniobra, triggers de bifurcación y condiciones de aborto. Genera el plan con el mejor modelo disponible (redactor), lo somete a…
Best practices for building with the Perigon News Intelligence API. Use this skill when searching or filtering news articles, tracking evolving stories, generating AI-powered summaries, performing semantic vector search over news or Wikipedia, querying people, companies, journalists, or sources databases, or…
Use when building, running, playtesting, or debugging a Godot 4.7 game through the NIUA Godot MCP (mcpniua-godot tools): create/open/save scenes, nodes, scripts, runmainscene / runcustomscene / runplaytestevidence, runtime probe, screenshots. Also use for untitled or unsaved scenes, nomainscene, stuck Save As or…
Scaffold production-ready Claude Code skills from a one-sentence description. Use when creating a new skill, bootstrapping a skill project, or turning an idea into a published skill. Generates SKILL.md, README.md, and domain-specific templates with best practices baked in.
An automated personal CRM, or customer and relationship manager, that extracts contacts and interactions from Gmail, calendars, and meeting notes. It maintains relationship summaries and health scores, sends follow-up reminders and meeting briefs, and drafts follow-up messages.
Iteratively reviews recent code changes (a PR if one exists for the current branch, otherwise the branch's commit range) and fixes what the reviews find. Runs in-depth-review roles behind a workflow barrier each iteration (gh-style-review too, in iteration 1, when invoked with --gh-style), merges and deduplicates…
Share reusable agentic patterns from project repos back into an upstream template, safely. WHEN TO PROPOSE: (1) After pushing commits touching high-signal paths, (2) Sprint/milestone review, (3) When agentic tooling is created. Classify commits as: YES (portable), MAYBE (needs changes/judgment—offer to fix), or NO…
Write features, fix bugs, and create PRs for the NanoClawbster codebase. Works in an isolated dev workspace, pushes via Composio (CI validates on GitHub Actions), and deploys after user approval.
AEC building modeling skill — parametric architectural geometry for Agent harness. USE when the task involves creating building geometry (floors, columns, walls, curtain walls, stairs), running code compliance checks (GB 50016, JGJ/T 67), exporting to STEP/IFC/GLB, or orchestrating multi-step building design workflows…
Cleans up local Git worktrees and branches whose work is provably absorbed elsewhere, deleting each risk group only after its own confirmation. Use after merging pull requests or after parallel agent worktree sessions.
A document-driven system for managing ongoing software projects. It keeps structured project documents so an AI can understand the current state, suggest and make changes, review results, and record what was learned.
Create new skills, modify and improve existing skills, and measure skill performance. Use when users want to create a skill from scratch, edit, or optimize an existing skill, run evals to test a skill, benchmark skill performance with variance analysis, or optimize a skill's description for better triggering accuracy.
A project-planning skill that turns a vague software idea into a structured CONTEXT.md blueprint. The blueprint describes the system, chosen technologies, data, APIs, and development milestones.
Debug a failed dynobox run from its terminal output. Use this skill when the user pastes or references output from npx dynobox run (especially --debug output with failed assertions), asks why a dyno run, scenario, or assertion failed, or wants a failing dyno fixed. Produces a root-cause diagnosis and proposed fixes…
Expert knowledge of rcurses Ruby TUI library for creating terminal user interfaces. Activate when building Ruby TUI applications, working with terminal panes, handling user input, or debugging rcurses applications. Covers initialization, pane management, text formatting, cursor control, and common pitfalls.
A full-stack programming and software design assistant, described in Chinese, for new and existing code projects.
★not rated 2 8mo agoA68 tokens
At most 3 mods per repository are shown here, and a mod shipped inside a plugin is left to that plugin's page — the rest are on their repository pages: