agent-arche

Ultra-compressed communication mode. Cuts output tokens 65% (measured) by speaking like caveman while keeping full technical accuracy. Supports intensity levels: lite, full (default), ultra, wenyan-lite, wenyan-full, wenyan-ultra. Use when user says "caveman mode", "talk like caveman", "use caveman", "less tokens", "be brief", or invokes $caveman. Also auto-triggers when token efficiency is requested.

Review a branch, PR, commit range, or working-tree diff along two separate axes: Standards and Spec alignment. Use for review requests and as the quality gate after non-trivial implementation. Load security review only for security-sensitive changes.

Shared vocabulary for designing deep modules. Use when the user wants to design or improve a module's interface, find deepening opportunities, decide where a seam goes, make code more testable or AI-navigable, or when another skill needs the deep-module vocabulary.

Diagnosis loop for hard bugs and performance regressions. Use when the user says "diagnose"/"debug this", or reports something broken/throwing/failing/slow.

Compact the current conversation into a handoff document for another agent to pick up.

Implement requested code changes end to end from a prompt, spec, or ticket. Use for feature work, diagnosed bug fixes, refactors, and other tasks that require editing code and verifying the result.

Scan a codebase for deepening opportunities, present them as a visual HTML report, then grill through whichever one you pick.

Behavioral guidelines to reduce common LLM coding mistakes. Use when writing, reviewing, or refactoring code to avoid overcomplication, make surgical changes, surface assumptions, and define verifiable success criteria. Load by default for implementation and code review work.

PostgreSQL database patterns for query optimization, schema design, indexing, and security. Use only when PostgreSQL is confirmed in the codebase or explicitly named by the user, and the task involves SQL, migrations, schemas, slow queries, Row Level Security, indexes, or connection pooling. Do not trigger for generic data work or a different database. Based on Supabase-leaning Postgres practice.

First-run setup for agent-arche projects. Analyze a codebase, seed repo memory when the installed workspace includes it, and configure docs/agents issue-tracker, triage-label, and domain context used by chained engineering skills. Use after installation or when that context is missing; in Skills-only workspaces, skip memory and configure only the available project context.

Threat-model, test, and review security-sensitive code changes. Use for authentication, authorization, sessions, secrets, untrusted input, API routes, SQL, file access, uploads, redirects, SSRF-prone outbound requests, serialization, cryptography, payments, or sensitive data; also use when the user requests an OWASP or vulnerability audit. Do not trigger for unrelated code merely because security is generally important.

Implement or audit search-specific website behavior: crawlability, sitemap and canonical rules, page metadata, structured data, search performance, Core Web Vitals, and internal linking. Use only when SEO/search visibility is requested or directly affected; do not trigger for generic frontend or performance work.

Test-driven development. Run a focused red-green loop for testable behavior changes and regression fixes. Use when the user requests test-first work, when `implement` identifies a useful public seam, or when `diagnosing-bugs` has a reproducible failure. Do not trigger for docs, configuration-only edits, generated files, or changes with no meaningful executable behavior.

Turn the current conversation into a spec and publish it to the project issue tracker — no interview, just synthesis of what you've already discussed.

Break a plan, spec, or the current conversation into a set of tracer-bullet tickets, each declaring its blocking edges, published to the configured tracker — edges as text in one file per ticket locally, or native blocking links on a real tracker.

Consolidated UI/UX design skill for product-fit direction, new interface implementation, redesigns, animation and micro-interactions, GSAP/ScrollTrigger choreography, style modes such as soft/minimalist/brutalist, UX critique, UI audits, UI performance optimization, Google Stitch DESIGN.md generation, and complete output enforcement. Use for any request naming design-intelligence, redesign, animate, gsap, soft, minimalist, brutalist, stitch, critique, ui-audit, ui-optimize, output, or any UI/UX/frontend visual work.

Consolidated Git workflow skill. Use when creating or reviewing branches, commits, pull requests, PR descriptions, squash commits, or release-ready Git hygiene. Loads branch, commit, and PR standards from steps/.

Grilling session that challenges your plan against the existing domain model, sharpens terminology, and updates documentation (CONTEXT.md, ADRs) inline as decisions crystallise. Use when the project has already started and the user wants to stress-test a plan against code, domain language, or documented decisions.