'Stage files and create a conventional commit with a well-formed message.'

Design product behavior and user experience with a product designer mindset. Use when shaping features, reviewing UX, improving flows, evaluating usability, clarifying copy, or making user-facing trade-off decisions.

Enforce the phase implementation workflow for Palette Kit v0.3 phases. Use when implementing a roadmap phase, creating a phase branch, validating the lint/typecheck/test/build pipeline, generating a patch for review, or preparing commit/push/PR steps.

>-

Draft, rewrite, critique, or analyze written communication in a reserved, pragmatic software-engineering voice. Routes internally to publication guidance in references/ for direct messages, PR reviews, internal and external email, blog posts, memos and broadcasts, technical documentation, RFCs and design docs, plus style analysis. Use when channel fit, tone control, structure, and writing quality matter. Do not use for intentionally hype-driven or playful marketing copy unless the user explicitly asks for it.

Expert guide for the JJ DOM manipulation library. Load this skill whenever you need to write, debug, or review JJ code; create native web components using JJ's defineComponent, setShadow, fetchTemplate, or fetchStyle; translate React, Vue, Svelte, Angular, jQuery, or Lit patterns to JJ idioms; work with JJHE, JJD, JJSE, JJME, JJDF, JJSR, JJET, JJN, or JJT wrappers; or write JJ tests with jsdom. If any JJ class name or helper function appears in the conversation, always load this skill.

Create a GitHub pull request using gh CLI with proper formatting

Load the shared GO execution workflow from `.agent/skills/go/shared/go.md`.

Session-end review of recent chiron activity. Scores across 5 axes (design thinking, code quality, idioms, testing, engineering maturity) and names one concrete thing to practice next time. Read-only in v0.3.0 — does not persist scores.

Address ALL PR issues (GitHub Actions, Greptile, SonarCloud, CI/CD)

Provide governance context for defining and maintaining boundaries between `*.action.ts` and `*.service.ts` in Next.js App Router entity APIs. Use when handling entity API layer functions or Server Action/Service methods, and when creating/refining conventions, refactor plans, or review criteria in `src/**/api`.

Culture assessment and monitoring with three modes — Diagnostic for a deep session surfacing the gap between declared culture and actual culture, Review for periodic check-ins on whether behavior is shifting, and Alert Triggers for setting tripwires around culture-strategy misalignment, regression, and integration risk. Not a culture change programme — a mirror. Updates company-context.md after every run. Triggers on questions about culture, values, team dynamics, how decisions really get made, why change efforts stall, merger integration, or whether the organisation can actually execute its strategy. Based on practice by Eterdis (eterdis.com).

Continuous environmental scanning across six PESTEL categories — Political, Economic, Social, Technological, Environmental, Legal. Not a one-time analysis but an ongoing radar. Three modes. Diagnostic mode runs a full scan and prioritises forces by impact, speed, and reversibility. Review mode checks the watch list monthly for movement and new signals. Alert mode defines concrete monitoring triggers for each top force. Use whenever someone wants to understand external forces acting on their business, assess macro risk, prepare for planning, or check whether the world has moved since last time they looked. Connects to the strategy map (forces act on the chain) and Wardley Map (forces drive component evolution). Based on practice by Eterdis (eterdis.com).

First-principles strategy — strip away conventional assumptions and find the theoretical maximum for your business. Three modes. Diagnostic mode runs a full session (45+ min) to define your theoretical maximum, map the gap honestly, classify constraints, invert for failure modes, set strategic intent, and work backwards to your first intervention. Review mode (15-20 min) checks whether your theoretical maximum has shifted, whether the gap has changed, and whether you're still climbing the right peak. Alert mode defines constraint-shift, peak-shift, and local-peak triggers that tell you when to reassess. Use when someone is stuck optimising incrementally, when conventional wisdom is driving strategy, when a new technology or regulation changes what's physically possible, or when you suspect the business is climbing a local peak. Integrates with company-context.md. Based on practice by Eterdis (eterdis.com).

Stress-test plans before committing by imagining they've already failed. Three modes. Diagnostic mode runs a full pre-mortem session — vivid failure scene-setting, six failure categories, the monkey test, and concrete plan adjustments. Review mode checks previous findings monthly for changed probabilities and new failure modes. Alert Triggers mode defines and monitors early warning signals across four trigger categories. Use before any major commitment — strategy, investment, product launch, partnership — or regularly to keep assumptions honest. The planning fallacy is real. The spreadsheet looked great so everyone stopped thinking. This is the antidote. Based on practice by Eterdis (eterdis.com).

Track Two / internal innovation system for running small, fast experiments outside the core business. Runs in three modes — Diagnostic for designing the full Protoloop from scratch (intake channels, loop structure, team, exit paths), Review for running the monthly heartbeat meeting (loop-by-loop decisions, forward-loading, portfolio health), and Alert Triggers for defining tripwires that protect, feed, and maintain momentum in the system. Updates company-context.md after every run. Triggers on questions about innovation pipelines, intrapreneurship, Track Two strategy, experimentation systems, handling ideas that don't fit the core business, or building systematic exploration alongside exploitation. Based on 20 years of practice by Eterdis (eterdis.com).

Strategy map analysis using the Eterdis two-directional causal chain. Use this skill to diagnose whether strategy is coherent by building two chains — bottom-up from resources to results, top-down from expectations to requirements — and finding where they don't connect. Runs in three modes — full diagnostic for deep sessions, review mode for quick re-assessment, and alert triggers for early-warning signals. Updates company-context.md after every run. Triggers on questions about strategic alignment, why the strategy isn't working, what to invest in, how the business creates value, whether the plan is realistic, or where the biggest gap is. Based on practice by Eterdis (eterdis.com).

VRIO resource analysis combined with Dynamic Capabilities assessment. Use this skill to identify real competitive advantages (Barney VRIO framework), then test whether the organisation can sense new opportunities, seize them, and transform itself to stay ahead (Teece dynamic capabilities). Runs in three modes — full diagnostic for deep first-use sessions, review mode for quick re-assessment, and alert triggers for defining early-warning signals. Updates company-context.md after every run. Triggers on questions about competitive advantage, defensibility, resource assessment, why competitors are gaining ground, organisational agility, or whether advantages will hold. Based on practice by Eterdis (eterdis.com).

Use when the user asks to generate a blog cover image, thumbnail, or article header. Automatically uses modern typography, brand logos, and Google Search grounding to create beautiful 16:9 images with Gemini 3.1 Flash Image Preview.

UI design guidelines for Kubani web interfaces. Use when building or reviewing web components, pages, or the Kubani dashboard.

>

Performs comprehensive security and performance audits of codebases, identifying vulnerabilities, unsafe patterns, security weaknesses, and performance bottlenecks. Use this skill when the user requests a security assessment, penetration testing analysis, vulnerability scan, or performance review of their code. Generates detailed, actionable reports with severity classifications and remediation guidance.

Contains MANDATORY code quality rules enforced by CI. MUST be consulted before refactoring or when hitting ESLint limits. Covers complexity thresholds, extraction patterns, and rule suppression.

Analyze conversation history, find gaps and drift in AGENTS/CLAUDE instructions and skills, propose and apply targeted improvements.

Complete codebase audit combining code review, Prisma query optimization, and TypeScript analysis with intelligent routing. Use when asked for a full project review, complete codebase audit, or comprehensive quality check of a full-stack application.

Implement code changes according to a written specification.

Async via gask, end turn immediately; use only when user explicitly delegates to Gemini (ask/@gemini/let gemini/review); NOT for questions about Gemini itself.

Interact with GitHub via the gh CLI for PRs, issues, releases, Actions, and API calls. Use when performing any GitHub operation such as creating PRs, viewing issues, checking CI status, making API requests, or managing releases.

GitHub CLI (gh) reference for authentication, repos, pull requests, issues, code review, Actions, releases, gists, API calls, search, aliases, extensions, codespaces, and configuration. Use when the user asks to create or manage PRs, issues, releases, workflows, gists, repos, or any GitHub operation from the terminal.

Discover FRESH issues in WELL-MAINTAINED repos (200+ stars). Merge-optimized: 60% easy wins (docs, typos, tests) + 40% bug fixes. Target agentic AI repos by CRITERIA (topic:llm/agent/rag + stars:>200). Verify repo health before queuing.

Apply these principles when writing, reviewing, or refactoring code. They are organized from the most fundamental (naming) to code organization, formatting, and commenting best practices.

Generate and edit images using Google's Gemini API (gemini-3-pro-image-preview model). Use when users request (1) Generating images from text prompts, (2) Editing existing images with AI instructions, (3) Creating images with specific styles or templates, (4) Generating multiple variations of images, (5) Creating images with reference images for style consistency, (6) Any image generation task mentioning Gemini, Google AI, or requiring professional image output. Supports aspect ratios (1:1, 3:4, 4:3, 4:5, 5:4, 9:16, 16:9, 21:9), 7 style templates (3 glass styles, 4 other creative styles), reference images (up to 14), and batch processing.

>

crusty-old-engineer

Review changed code for reuse, quality, and efficiency, then fix any issues found.

Use when about to orchestrate a subagent-driven-development execute-plan session — provides 5 realistic task scenarios with Amplifier delegate() patterns, model_role selection, status handling (DONE, DONE_WITH_CONCERNS, NEEDS_CONTEXT), and fix loops for spec and quality issues

Complete reference tables for Superpowers modes, agents, recipes, and anti-patterns

- Security Architect

Search through your subagents and skills and make a detailed plan to execute the user's query. Use the subagents to save context and use shared documents in agent_tasks as a form of memory and passive

Problem definition → small, safe change → change review → refactor — repeat the loop.

This scope governs the Claude Code subagent wiring files that connect Claude Code's Agent tool to Armature persona definitions. Core agent wiring (reviewer, planner, red team) uses thin pointers to `.

End-of-week review. Summarize the week, surface patterns, prune stale notes.

End-of-session workflow. Summarize the session, update wiki, write log.

plan-ceo-review

Operate Microsoft Visio through the DrawForge bridge with stable atomic actions. Use when an agent needs to create sessions, add/edit/connect/style shapes, save documents, export PNG previews, or execute a prepared drawing spec without taking over planning decisions.

- `src/` — TypeScript sources for the composite action (`dependency-review.ts`, `check-ossf.ts`, `check-guarddog.ts`, `frozen-install.ts`).

Create or update GitHub pull requests using gh CLI. Use when creating PRs, updating PR titles/descriptions, opening pull requests, or submitting changes for review.

以资深架构师视角对 Diff/源码做深度 Code Review：正确性与防御性、架构合规、DRY、KISS、测试完备性；禁止格式挑刺；单次最多 5 条高价值问题；无重大问题输出 LGTM。在用户附上改动、请求审查、PR review、走查、diff 评审时触发。

Enforces Next.js 16 best practices for App Router, React Server Components, caching, routing, TypeScript, and performance. Use when building, reviewing, or refactoring Next.js applications, creating pages/layouts/routes, configuring next.config.ts, handling data fetching, or working with server/client components.

Create a GitHub Pull Request to merge the current branch into main with a well-structured title and description using GitHub CLI. Use when the user asks to create a PR, open a pull request, submit a merge request, or send changes for review.