“refactoring”

Drive a Claude Code session toward a verifiable goal until done or a hard cap fires - autonomous loop, no user intervention.

Use when a project needs AI agent rules built or upgraded — no AI_RULES.md exists, an existing rule index has grown into one bloated file, or the user asks…

Create visual, intuitive implementation plans for coding tasks.

The definitive source of truth for all coding conventions, environment constraints, and architecture layouts in the SICTIC-AI repository.

Test-driven development.

Run the `fallow` CLI on any JavaScript/TypeScript project to find dead code, unused dependencies, duplication, complexity hotspots, and security candidates…

MANDATORY CORE SKILL: Enforces the Autonomous Coding & Self-Correction Loop for Antigravity in every session.

Personal working mode for software engineering tasks.

Scan codebases for technical debt with AST parsing, prioritize by impact, and generate trend dashboards.

Converts user assumptions, feature requests, or bug reports into verified, test-driven code changes.

Reshape the commits a branch carries on top of its target branch into a clean, logical sequence without changing the final committed tree.

Independent peer review of an implemented change (the diff) via a fresh subagent: a reviewer with no memory of the implementation session pressure-tests the…

MUST USE for ANY work on .py .pyi .rs .ts .tsx .mts .cts .go files.

Run Go static analysis + complexity checks (gocognit, gocyclo, staticcheck).

Make changed code easy for agents and humans to find, understand, and verify through search-oriented naming, precise types, truthful local guidance, coherent…

Guide maintainable, resilient, test-double-friendly code: clear names, decomposition, safe defaults, idempotency, bounded retries, observability, and quality…

Detects and safely removes unused code (imports, functions, classes) across multiple languages.

NestJS best practices and architecture patterns for building production-ready applications.

Orchestrates any non-trivial development or analysis task end-to-end — discovery, alignment, blueprint, registration, implementation and delivery.

Use when tasks need AI multi-agent factor mining research boundaries, versioned ResearchBrief/FactorSpec contracts, evaluation/review/decision objects, or…

Cherry-pick, backport, or apply commits/PRs onto another branch with safety gates and per-change validation; also release audits — "what's on master that…

poteto's agent style for concise, detailed responses, deliberate subagents, unslopped prose, simple code, and verified work.

Measures project architecture and identifies files worth reviewing with concrete hints.

Plan and write Playwright end-to-end coverage for user-facing generated application features, flows, and failure states.