Claude Skills by big-emotion

Production-readiness audit for Big Emotion Project Standard (this repo). Read-only multi-domain scored assessment that answers four questions — is the plugin ready to install on real repos, is the template surface healthy, what is the security posture, and is the score close to 8–9/10. Use when the user asks "is it ready", "audit the project", "production-readiness check", or invokes /project-standard-audit.

One-shot bootstrap of the Confluence spec tree (Requirements / Decisions / Architecture / Obsolete) for Big Emotion Project Standard. Creates the four skeleton subpages under the project's engineering root page, records their IDs in the spec config, writes the sentinel, then refuses second runs. Use only when initializing the spec system, never for ongoing maintenance.

Prepare and ship an Agent Atelier release. Bumps the semver version in the lockstep manifests (package.json, every plugins/*/.claude-plugin/plugin.json, every .claude-plugin/marketplace.json plugins[] entry), updates CHANGELOG.md (Keep a Changelog format, creates it if missing), creates an annotated git tag on main, then asks for explicit confirmation before pushing. Tag push triggers no workflow — this repo has no deploy; the GitHub Release is created directly by this skill. Use when the use...

Spec maintainer for Big Emotion Project Standard. Reads Confluence (Requirements / Decisions / Architecture) as the source of truth, helps you investigate a problem, structures the reflection, and produces drafts — Pending REQ/DEC/ARCH sections on Confluence + matching Jira tickets in the Jira project configured in docs/confluence-spec/config.json, using the project template. Append-only on Confluence, Pending only. Use when you face a bug, want to add a feature, or need to refine an idea.

End-to-end local automation for a single Jira ticket on the Big Emotion Project Standard repo. Paste a Jira ticket URL (or key) and the skill self-assigns the ticket, reads it, refines it, creates the Jira sub-tasks, branches off the integration branch in an isolated git worktree, implements the work in parallel via sub-agents, opens the pull request, then moves the ticket to the review column and comments the PR link. Runs fully automatically with no confirmation gates. Use when the user pas...

Scan a code diff for the consistent "tells" of unreviewed AI-generated code — narrating comments, generic names, hollow over-mocked tests, signature-restating docs, and unjustified decisions. Read-only; reports each finding with file:line, why it's a tell, and the concrete fix. Use when the user says "check for AI tells", "review this diff for AI smells", "ai-code-tells", or invokes /ai-code-tells. Works standalone; pairs well with any always-on code-quality rules you define in your own CLAUD...

Produce a specs/ directory of flat markdown files from any frontend codebase. Covers 12 axes needed for rebuild/migration. Use when the user asks to "spec a codebase", "create a blueprint", "extract specs for migration", or wants to document a frontend app for rebuild.

Reverse-engineer any web application into a complete pixel-perfect reproduction kit. Extracts design tokens, DOM structure, text content, assets, and responsive screenshots across multiple pages. Use when the user asks to "reverse-engineer a UI", "extract a design system", "analyze the design of a website", or wants to reproduce a site pixel-perfect.

Validate specs and implementations against the reverse-UI oracle. Catches drift before it ships. Use when checking if a spec accurately describes the original UI or if an implementation matches the original.

Batch-fix the CI on every open pull request whose required checks are red — CI only, no review, no merge. From a GitHub PRs URL (or the current repo), enumerate every open PR with failing required checks, fetch each PR's failing-check logs, consolidate a per-PR CI-fix checklist, dispatch a team of parallel fixer sub-agents (one isolated worktree per PR) that apply the minimal fix, run the repo's quality gates locally, push to each PR branch, then poll CI until it turns green (or a break-out c...

Batch-apply the requested changes on every open pull request in a GitHub list — fix only, no merge. From a GitHub PRs URL (or the current repo), enumerate every open PR, collect each PR's requested changes (formal CHANGES_REQUESTED reviews, inline review comments, and `<!-- review-train -->` "Issues requiring changes" comments / `changes-requested` label), consolidate a per-PR fix checklist, dispatch a team of parallel fixer sub-agents (one isolated worktree per PR) that implement the minimal...

Batch-process every open GitHub issue in a list — triage, detect existing PR linkage, detect silent fixes in already-merged PRs, close issues that are no longer pertinent, then plan + build + open one PR per remaining actionable issue with `Closes #<n>` in the body so GitHub's Development field links the PR to the issue and merging the PR auto-closes it. From a GitHub issues URL (or the current repo), enumerate every open issue, dispatch a team of parallel triage sub-agents (one per issue) to...

Batch-process a list of open GitHub pull requests. From a GitHub PRs URL (or the current repo), enumerate every open PR, compute the safe merge order from dependencies, then per PR fix CI, resolve merge conflicts, run a code review, and assign a status (mergeable / changes-required / blocked-by-order). Presents one consolidated plan, asks for a single grouped confirmation, then fixes + merges every mergeable PR in order and posts an order-blocking status comment on the rest. Generic and repo-...

Batch code-review every open pull request in a GitHub list — review only, nothing else. From a GitHub PRs URL (or the current repo), enumerate every open PR, dispatch a team of parallel reviewer sub-agents (one per PR), and have each produce a high-detail five-axis review in the exact structured format of the project's reference review. Presents one read-only plan, asks for a single grouped confirmation, then posts one review comment per PR and applies an `approved` or `changes-requested` lab...

Install and configure the Big Emotion project standard on a new or existing repo — CI quality gates, Husky hooks, the four project skills, Atlassian wiring (Jira board + Confluence spec tree), Ferry, and the infrastructure module (OVH VPS deploy, Azure variant, M365 mail, secrets doctrine). Use when starting a new Big Emotion project, bringing an existing repo up to standard, checking a repo's compliance, or setting up deploy/mail/secrets ("setup this project", "install the standard", "gap an...

Production-readiness audit for {{project_display_name}} (this repo). Read-only multi-domain scored assessment that answers four questions — is the project ready for production, is its domain-critical surface healthy, what is the security posture, and is the score close to 8–9/10. Use when the user asks "is it ready", "audit the project", "production-readiness check", or invokes /{{project_slug}}-audit.

One-shot bootstrap of the Confluence spec tree (Requirements / Decisions / Architecture / Obsolete) for {{project_display_name}}. Creates the four skeleton subpages under the project's engineering root page, records their IDs in the spec config, writes the sentinel, then refuses second runs. Use only when initializing the spec system, never for ongoing maintenance.

Prepare and ship a {{project_display_name}} production release. Bumps the semver version of the root package.json, updates CHANGELOG.md (Keep a Changelog format, creates it if missing), creates an annotated git tag on {{default_branch}}, then asks for explicit confirmation before pushing. Tag push triggers {{deploy_workflow}}. Use when the user says "release {{project_slug}}", "cut a release", "bump version", "tag a new version", or invokes /{{project_slug}}-release.

Spec maintainer for {{project_display_name}}. Reads Confluence (Requirements / Decisions / Architecture) as the source of truth, helps you investigate a problem, structures the reflection, and produces drafts — Pending REQ/DEC/ARCH sections on Confluence + matching Jira tickets in {{jira_project_key}} with the project template. Append-only on Confluence, Pending only. Use when you face a bug, want to add a feature, or need to refine an idea.

End-to-end local automation for a single Jira ticket on the {{project_display_name}} repo. Paste a Jira ticket URL (or key) and the skill self-assigns the ticket, reads it, refines it, creates the Jira sub-tasks, branches off the integration branch in an isolated git worktree, implements the work in parallel via sub-agents, opens the pull request, then moves the ticket to the review column and comments the PR link. Runs fully automatically with no confirmation gates. Use when the user pastes ...

Identify the precise professional role behind a topic or rough prompt, map lay wording to the field's proper terminology, and rewrite the prompt as an expert-grade prompt prefixed with "You are a <role>, and you will <mission>". Use when the user invokes /expertify, asks "what's the proper term for…", "what do you call the person/role who…", or wants their prompt upgraded with the right domain vocabulary before asking it.

Analyze a GitHub repo to understand what it does, extract its best parts, find reusable components, and get actionable usage instructions. Use when the user shares a GitHub URL and wants a practical breakdown.

Capture a surprising model answer as a thought entry. Synthesizes the last answer with context, saves as a granular file, and auto-updates the TLDR.