Agentsys

Master workflow orchestrator: autonomous workflow with model optimization (opus/sonnet/haiku), two-file state management, workflow enforcement gates, 8 specialist agents

Pre-ship quality gates: deslop, simplify, agnix, enhance, review loop, delivery validation, docs sync

Quality gates then ship - chains /prepare-delivery then /ship in one command

Complete PR workflow: commit to production, skips review when called from next-task, removes task from registry on cleanup, automatic rollback

3-phase AI slop detection: regex patterns (HIGH), multi-pass analyzers (MEDIUM), CLI tools (LOW)

Multi-agent iterative code review until zero issues remain

Deep repository analysis to realign project plans with code reality - detects drift, gaps, and creates prioritized reconstruction plans

Master enhancement orchestrator: parallel analyzer execution for plugins, agents, docs, CLAUDE.md, and prompts with unified reporting

Canonical guidance for writing, reviewing, and maintaining production-grade SKILL.md files across agent tools

Production-grade system prompt curation for autonomous coding agents and multi-agent orchestration

Standalone documentation sync: find outdated refs, update CHANGELOG, flag stale examples based on code changes

Unified static analysis via agent-analyzer - git history, AST symbols, project metadata, and doc-code sync

Rigorous performance investigation workflow with baselines, profiling, hypotheses, and evidence-backed decisions

Research topics online and create comprehensive learning guides with RAG-optimized indexes

Durable negative behavior memory: persist user-defined banned agent behaviors into CLAUDE.md or AGENTS.md

Lint agent configuration files (SKILL.md, CLAUDE.md, hooks, MCP) against 414 rules across 10+ AI tools

Cross-tool AI consultation: get second opinions from Gemini CLI, Codex CLI, Claude Code, OpenCode, or Copilot CLI with model and thinking effort control

Structured multi-round debate between AI tools with proposer/challenger roles and verdict

Learn from workflow patterns across sessions and suggest skills, hooks, and agents to automate repetitive work

Codebase onboarding - automated data collection and interactive project orientation

Find where to contribute to any project - matches developer skills to test gaps, stale docs, bugspots, and open issues

Zig language server for Claude Code via ZLS - automatic diagnostics after every edit, jump-to-definition, find-references, and hover. Requires zls in PATH

Teach agents to write idiomatic, correct, performant, current Mojo (v1.0.0b1) - syntax, ownership and CPU/memory optimization (copies, SIMD, vectorize/parallelize), GPU kernels (DeviceContext/LayoutTensor), and Mojo/Python interop; prevents stale pre-2025 syntax

Teach agents to write idiomatic, correct, current Ada and SPARK (Ada 2022) - contracts (aspects, Pre'Class), the Alire ecosystem, SPARK proof (AoRTE, assurance levels, ownership/borrow), embedded (Ravenscar/Jorvik), and GNAT/GNAT SAS tooling; blocks stale pre-2022 advice (GNAT Community, pragma contracts, CodePeer)

Reconstruct the real history of code (goal, shipped vs in-flight, coupled files) from Codex, Claude Code, and Eigen transcripts, surfaced via the get-oriented skill - so agents stop guessing the story of code they didn't write this session