selfagency

Use this skill whenever reviewing, auditing, or grading a command-line tool for agent-friendliness - it runs a black-box test suite against a target CLI and reports per-rule pass/fail from the cli-for-agents 45-rule catalog. Trigger even if the user doesn't explicitly say "agent-friendly" - apply whenever they ask "is mycli good for agents?", "review this CLI", "grade my cli against the rules", "check if this tool is safe to automate", or "audit command-line design". Companion to the cli-for-agents distillation skill.

Conduct comprehensive code reviews addressing syntactical and programmatic correctness, data flow, type safety (zero `any`), performance, security, API conformity, documentation consistency, best practices, and code style. Produces an exhaustively detailed remediation plan before implementing any fixes.

Use this skill for repositories managed through the git-mcp server when the task involves Git inspection, staging or committing, branches, remotes, rebase, cherry-pick, stash, bisect, worktrees, Git Flow, LFS, release tagging, PR preparation, or recovery. Prefer git-mcp tools over shell git commands, including when the user asks in raw Git terms such as git_status, git_commit, git_push, git_rebase, git_reflog, git_flow, submodules, or “undo this safely”.

Convert ANSI art composition principles into Ink UI chrome, layout, and component styling.

Design React Ink component structures, props, and state for a terminal IDE with BBS chrome.

Validate keyboard flow, focus behavior, resize behavior, ANSI readability, and terminal usability for Ink BBS-style UIs.

Design React Ink terminal UI components for a terminal-based IDE with classic BBS / warez-scene aesthetics and ANSI art composition rules.

Design React Ink terminal UI components for a terminal-based IDE with classic BBS / warez-scene aesthetics and ANSI art composition rules.

Install and configure JavaScript/TypeScript typechecking, linting, formatting, Git pre-commit hooks, and GitHub Actions CI checks. Use this skill when the user asks to set up or migrate code quality tooling with Husky and lint-staged, choose between Oxc (oxlint+oxfmt), Biome, or ESLint+Prettier, optionally select Rumdl as Markdown linter/formatter, add `tsc --noEmit` typechecking, configure GitHub Actions CI for check-only validation, copy existing config files, initialize configs with official tool commands, or enforce either check-only or auto-fix pre-commit behavior for TS/JS/YAML/JSON/Markdown files.

Use this skill when setting up 1Password Environments for local development to mount named secret sets as local .env files via FIFO pipes, configuring the .1password/environments.toml file with mount paths, installing agent hooks to validate environment files before AI coding assistant execution (Claude Code, Cursor, GitHub Copilot, Windsurf), or troubleshooting FIFO .env delivery issues. Not for secrets automation or SSH key management.

Use this skill to securely integrate 1Password secrets into codebases, AI agents, or CI/CD pipelines using secret references (op://vault/item/field), the op run --env-file pattern, 1Password SDK secrets.resolve() for AI agents with service accounts, MCP server wrapping with op run to avoid plaintext secrets in mcp.json, or the GitHub Actions load-secrets-action and CircleCI secrets orb. Also use to audit a codebase for hardcoded secrets.

Set up and use 1Password CLI (op). Use when installing the CLI, enabling desktop app integration, signing in (single or multi-account), reading/injecting/running secrets via op run or op inject, configuring shell plugins for biometric auth of CLIs (AWS, GitHub, etc.), or creating and using service accounts for programmatic vault access.

Use this skill when setting up the 1Password SSH agent, generating or importing SSH keys into 1Password, configuring SSH clients to authenticate through 1Password (IdentityAgent, SSH_AUTH_SOCK), setting up Git commit signing with SSH keys from 1Password, configuring SSH agent forwarding for cloud development environments or remote workstations, or troubleshooting SSH authentication failures. Private keys never leave 1Password.

Automate system administration workflows on macOS and Linux. Use this skill when: setting up reproducible dev environments (Brewfile), parsing CLI command output programmatically (jc + jq), monitoring remote servers via SSH with alerts, integrating API keys and secrets securely (1Password CLI), building automation workflows (Cherri Siri Shortcuts programming), or batch-managing system configurations. Covers package management (Homebrew), structured JSON parsing (jc), SSH-based remote administration, secure credential injection, cross-platform multi-step automation, and fleet monitoring.

Write advanced shell automation scripts with Google zx, a Node.js tool that wraps subprocess execution with ergonomic template literals. Use when asked to write a shell script, bash script, or automation script in Node.js; automate CLI workflows; spawn subprocesses; pipe commands; process command output; replace a bash script with modern JavaScript; write cross-platform automation; run terminal commands programmatically; or when the user mentions zx, $`command`, shell scripting, subprocess management, or Node.js scripting. Covers: subprocess execution, piping, parallel execution, error handling with nothrow and retry, streaming output, jc JSON parsing integration, and shell one-liner idioms translated to zx.

Use this skill when a user needs to inspect, troubleshoot, deploy, restart, or manage resources in a Coolify environment (self-hosted or Coolify Cloud) via the @masonator/coolify-mcp server. Trigger for requests about Coolify applications, databases, services, servers, projects, environments, deployments, environment variables, backups, private keys, GitHub app integrations, teams, or cloud tokens—even if the user does not explicitly say “Coolify” (for example: “redeploy my API”, “why is staging unhealthy?”, “restart postgres”, “what’s broken in my infra?”). Prefer Coolify MCP tools over direct API calls, curl, or ad-hoc shell commands when available.

Use this skill when the user needs TeamDynamix ITSM operations through MCP: ticket triage and updates, KB article maintenance, asset/CI relationships, service catalog lookups, project issue/risk workflows, or time entry queries. Always discover IDs first, use read-before-write patterns, and enforce write safety gates.

Use this skill when the user needs TeamDynamix ITSM operations through MCP: ticket triage and updates, KB article maintenance, asset/CI relationships, service catalog lookups, project issue/risk workflows, or time entry queries. Always discover IDs first, use read-before-write patterns, and enforce write safety gates.

Generate a complete MCP server project in TypeScript with tools, resources, and proper configuration

Implement and manage Stately stores — the Pinia-inspired Svelte 5 state management library (@selfagency/stately). Use when creating defineStore() option or setup stores, configuring plugins (persistence, history, FSM, sync, async, validation), wiring SSR-safe managers in SvelteKit, using storeToRefs(), $patch, $subscribe, $onAction, or any Stately API. Also use when debugging store reactivity, choosing concurrency policies, modeling finite state machines, adding time-travel, or building custom Stately plugins. Triggers on: "create a store", "add persistence", "add history/undo/redo", "sync across tabs", "finite state machine", "async plugin", "validation plugin", "SSR-safe store", "SvelteKit data loading with stores", "storeToRefs", "$patch", "$subscribe", "defineStore", "createStateManager", "stately".

Use for Beans issue tracker workflows in this workspace: planning and decomposing epics, listing issues, finding top-priority or stale work, starting work on a bean, creating/updating issues via extension commands or MCP tools, managing status/type/priority/parent/blocking relationships, and suggesting issue-related branch, commit, and PR workflows. Triggers on: "create a bean", "what should I work on", "plan this epic", "decompose this feature", "find stale issues", "set priority", "start work", "what beans are in progress", "commit this work".

Configure and run WordPress-to-11ty migrations using Unpress. Use when migrating a WordPress site to static HTML, setting up unpress.yml project config, configuring media handling via local download or S3/SFTP reupload, creating .env files for WordPress credentials, enabling Meilisearch search indexing, troubleshooting migration failures, or customizing generated 11ty site output. Also use when the user mentions unpress, wordpress-to-11ty, wp-export, site generation, or media reupload.

Use this skill when the user needs to manage Beans work items in a Beans workspace—create, view, update, reopen, archive, query, or bulk-manage beans, edit `.beans` files/frontmatter, or manage parent/blocking relationships. Use it even when the user says tickets/issues/backlog instead of “beans.” Do not use it for generic GitHub issue workflows that are not backed by Beans.

Use this skill for repositories managed through the git-mcp server when the task involves Git inspection, staging or committing, branches, remotes, rebase, cherry-pick, stash, bisect, worktrees, Git Flow, LFS, release tagging, PR preparation, or recovery. Prefer git-mcp tools over shell git commands, including when the user asks in raw Git terms such as git_status, git_commit, git_push, git_rebase, git_reflog, git_flow, submodules, or “undo this safely”.

Discover and safely use the git-mcp server for Git inspection and mutation workflows.

Use this skill when the user needs to manage Beans work items in a Beans workspace—create, view, update, reopen, archive, query, or bulk-manage beans, edit `.beans` files/frontmatter, or manage parent/blocking relationships. Use it even when the user says tickets/issues/backlog instead of “beans.” Do not use it for generic GitHub issue workflows that are not backed by Beans.

Use for repositories managed through the git-mcp server when an agent needs to inspect repository state, create commits, manage branches and remotes, rebase, cherry-pick, stash, bisect, work with worktrees, handle Git Flow or git-flow-next-style workflows, manage tags, recover from mistakes, or explain Git concepts using MCP tools instead of raw git CLI commands. Especially relevant for prompts about git_status, git_commit, git_push, git_rebase, git_cherry_pick, git_worktree, git_stash, git_reflog, git_flow, git_lfs, submodules, PR preparation, release tagging, and Git recovery. IMPORTANT: Never use shell/terminal to run git commands directly; always use git-mcp MCP tools instead.

Use when performing security audits, vulnerability assessments, penetration testing, or when asked to find security issues. Use when encountering code that handles user input, executes commands, reads files, or makes network requests.

Use for Beans issue tracker workflows in this workspace: planning and decomposing epics, listing issues, finding top-priority or stale work, starting work on a bean, creating/updating issues via extension commands or MCP tools, managing status/type/priority/parent/blocking relationships, and suggesting issue-related branch, commit, and PR workflows. Triggers on: "create a bean", "what should I work on", "plan this epic", "decompose this feature", "find stale issues", "set priority", "start work", "what beans are in progress", "commit this work".

Use for Beans issue tracker workflows in this workspace: planning and decomposing epics, listing issues, finding top-priority or stale work, starting work on a bean, creating/updating issues via extension commands or MCP tools, managing status/type/priority/parent/blocking relationships, and suggesting issue-related branch, commit, and PR workflows. Triggers on: "create a bean", "what should I work on", "plan this epic", "decompose this feature", "find stale issues", "set priority", "start work", "what beans are in progress", "commit this work".