By agent-sh
Teach coding agents to write idiomatic, correct, performant, current Mojo (Modular's language) - syntax, ownership and CPU/memory optimization, GPU kernels, and Mojo/Python interop. Targets Mojo v1.0.0b1 and prevents stale pre-2025 syntax.
A skill that teaches any coding agent to write idiomatic, correct, performant, current Mojo (Modular's language).
Mojo is young and fast-moving. The dominant failure mode for agents is emitting 2023-2024-era syntax that now warns or fails to compile. This plugin ships an always-on stale-to-current version map plus focused guidance for fundamentals, performance, and GPU - so agents target the current language (Mojo v1.0.0b1), not their training data.
def not fn, var/read/mut not owned/borrowed/inout, @fieldwise_init not @value, synthesized __init__(copy=)/__init__(take=) not __copyinit__/__moveinit__, TileTensor/LayoutTensor not NDBuffer, TestSuite not the removed mojo test command.comptime metaprogramming, and typed error handling.SIMD, vectorize/parallelize/elementwise, @always_inline, LayoutTensor, benchmarking.DeviceContext, kernels, the thread/block/grid model, shared memory, barrier(), warps, occupancy (NVIDIA / AMD / Apple).claude plugin marketplace add agent-sh/mojo
claude plugin install mojo
Or add it through the agentsys marketplace alongside the rest of the agent-sh ecosystem.
The skill activates automatically on Mojo work. Triggers include:
write a Mojo struct for a 3D vector with SIMD ops
port this Python function to Mojo
is this idiomatic Mojo?
write a GPU vector-add kernel in Mojo
why is my Mojo parallelize loop slow?
It does not activate for Python, Rust, C++, or Modular MAX serving/deployment config.
skills/mojo/SKILL.md the skill (version map + core rules + workflow + doc links)
.claude-plugin/ Claude Code plugin + marketplace manifests
.codex-plugin/ Codex plugin manifest
agnix . # validate skill + manifests (zero errors required)
claude plugin validate . # validate marketplace and plugin manifests
Mojo guidance is grounded in current upstream docs at mojolang.org/docs, not model memory - re-verify any stdlib API outside stabilized-API markers against the installed toolchain.
skill-curator - for writing and reviewing skillsagnix - the linter that validates this pluginMIT
Own this plugin?
Verify ownership to unlock analytics, metadata editing, and a verified badge. GitHub access is read-only (username + org membership).
Sign in to claimOwn this plugin?
Verify ownership to unlock analytics, metadata editing, and a verified badge. GitHub access is read-only (username + org membership).
Sign in to claimBased on adoption, maintenance, documentation, and repository signals. Not a security audit or endorsement.
npx claudepluginhub agent-sh/mojo --plugin mojoProfessional-grade slash commands for Claude Code with cross-platform support
Lint agent configuration files (SKILL.md, CLAUDE.md, hooks, MCP) against 414 rules across 10+ AI tools
Find where to contribute to any project - matches developer skills to test gaps, stale docs, bugspots, and open issues
Deep repository analysis to realign project plans with actual code reality - discovers drift, gaps, and produces prioritized reconstruction plans
Master workflow orchestrator with autonomous task-to-production automation, quality gates, and multi-agent review
Persistent file-based planning for AI coding agents. Crash-proof markdown plans (task_plan.md, findings.md, progress.md) that survive context loss and /clear, with an opt-in completion gate and multi-agent shared state. Manus-style. Works with Claude Code, Codex CLI, Cursor, Kiro, OpenCode and 60+ agents via the SKILL.md standard. Includes Arabic, German, Spanish, and Chinese (Simplified and Traditional).
Comprehensive skill pack with 66 specialized skills for full-stack developers: 12 language experts (Python, TypeScript, Go, Rust, C++, Swift, Kotlin, C#, PHP, Java, SQL, JavaScript), 10 backend frameworks, 6 frontend/mobile, plus infrastructure, DevOps, security, and testing. Features progressive disclosure architecture for 50% faster loading.
Consult multiple AI coding agents (Gemini, OpenAI, Grok, Perplexity, plus codex, antigravity, and grok CLIs when installed) to get diverse perspectives on coding problems
Design fluency for frontend development. 1 skill with 23 commands (/impeccable polish, /impeccable audit, /impeccable critique, etc.) and curated anti-pattern detection.
Lazy senior dev mode. Forces the simplest, shortest solution that actually works: YAGNI, stdlib first, no unrequested abstractions.
Behavioral guidelines to reduce common LLM coding mistakes, derived from Andrej Karpathy's observations on LLM coding pitfalls