sema: Command for Claude Code

.claude/commands/swarm-hybrid.md

swarm-hybrid is a command for Claude Code from sema-lisp/sema. It costs 0 tokens per session (1,608 once invoked), scanned A, original, MIT.

A method for asking a mix of language-design personas and technical specialists to analyze a problem and form a consensus. The mix changes depending on whether the problem concerns design, architecture, implementation, or several of these.

In plain words
What is it for?
Use it for questions about syntax, semantics, APIs, module boundaries, virtual-machine instructions, performance, and testing strategy in Sema.
Why use it?
It brings both user-facing design opinions and implementation-focused checks into decisions that may affect the whole language project.

Command for Claude Code

Written for Claude Code: $ARGUMENTS substitution.

This is sema-lisp/sema's own configuration. It tells Claude Code how to work on sema itself, so it is not a mod to install elsewhere. Copy it as a starting point and replace the rules that are about this project. Everything sema configures →

Reuse

Borrowing it

Nothing to install: this file belongs to sema-lisp/sema. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.

Copy the file
curl -O https://raw.githubusercontent.com/sema-lisp/sema/main/.claude/commands/swarm-hybrid.md
Clone the repo
git clone --depth 1 https://github.com/sema-lisp/sema

Made for: Claude Code.

Wrote this? Show the measurements

A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.

agentmods badge for swarm-hybrid

README.md
[![agentmods](https://agentmods.dev/badge/commands/sema-lisp/sema/swarm-hybrid/github.svg)](https://agentmods.dev/commands/sema-lisp/sema/swarm-hybrid)
Your own site
<a href="https://agentmods.dev/commands/sema-lisp/sema/swarm-hybrid"><img src="https://agentmods.dev/badge/commands/sema-lisp/sema/swarm-hybrid/github.svg" alt="Measured on agentmods" height="20"></a>

Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.

agentmods 80×15 button for swarm-hybrid

Your own site · 80×15
<a href="https://agentmods.dev/commands/sema-lisp/sema/swarm-hybrid"><img src="https://agentmods.dev/badge/commands/sema-lisp/sema/swarm-hybrid.svg" alt="Reviewed on agentmods" width="80" height="20"></a>
Per session 0 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 1,608 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. A grade says what 26 rules found in the file — not that it is safe.
Origin original No closer match found in the catalogue.
Token cost

What it costs to keep this loaded

Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.

ModelPer sessionOnce invoked
Fable 5.1 $0.00000 $0.01608
Opus 5 $0.00000 $0.00804
Sonnet 5 $0.00000 $0.00322
Haiku 4.5 $0.00000 $0.00161

Measured 9d ago against content hash 9048b2ad63e2, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-09, from the pricing page.

Security

Grade A, and why

swarm-hybrid scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 9d ago.

A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.

Nothing flagged

None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.

.claude/commands/swarm-hybrid.md · 159 lines

How it starts

The opening of the file, as written. The whole thing — 159 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Swarm Consensus — Hybrid (Personas + Roles)

You are orchestrating a multi-agent consensus analysis. You automatically select the right mix of PL designer personas and Sema-specific technical roles based on the problem type.

Step 1 — Classify the problem and determine N

Read the problem and classify it:

Problem Type Indicator Mix
Design "Should we...", syntax choices, semantics, API shape, language philosophy 70% personas, 30% roles
Architecture Crate boundaries, module system, dependency direction, state management 50% personas, 50% roles
Implementation VM opcodes, lowering, specific functions, performance, testing strategy 30% personas, 70% roles
Mixed Touches multiple levels Even split

Pick N based on complexity:

Complexity N When
Low 3 Single concern, reversible
Medium 5 Multi-factor, moderate stakes
High 7 Cross-cutting, hard to reverse
Critical 10 Strategic, deeply uncertain

State: problem type, chosen N, and the specific agents you're spawning (with reasoning for each pick).


Step 2 — Select agents and spawn in parallel

PL Designer Personas (for design tension)

Persona Philosophy Pick When
Guy Steele Languages must grow. User extensibility is non-negotiable. Module/extensibility/macro questions
Rich Hickey Simple != easy. Immutable data composes; mutable state entangles. State management, API surface, data modeling
Rob Pike Less is exponentially more. Orthogonal features. Resist feature creep. Feature scope, when to say no
Joe Armstrong Design around failure. Isolation. Let it crash. Error handling, agents, fault tolerance
Simon Peyton Jones Purity enables reasoning. Effects should be explicit. Evaluation order, type-safety, correctness
Matthew Flatt Modules enable DSLs. Phase separation is crucial. Module system, macro phases, DSL embedding
Gerald Sussman Pragmatic formalism. Empirical testing beats theory. Evaluation strategy, scoping, closure design
Paul Graham Code=data. Macros reshape the language. Flexibility wins. Macro systems, metaprogramming, syntax
Roberto Ierusalimschy Mechanisms not policies. Small core. Embed everywhere. VM design, embedding API, minimal core
Peter Norvig Data over formalism. Leverage AI scale. Pragmatic patterns. LLM integration, AI-native features
Kent Pitman Community shapes languages. Codify practice. Conditions > exceptions. Error handling, stdlib conventions
Bob Nystrom Show the implementation. No black-box magic. VM clarity, dual evaluator design
Brendan Eich Good design under real constraints beats perfect design. Shipping pragmatism, mixed paradigms
Guido van Rossum Readability is social. One obvious way. Least surprise. DX, error messages, naming, onboarding

Read the full file on GitHub · 159 lines

Changes

What this file has done since we first saw it

Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.

  1. 9d ago First seen · 159 lines · 0 tokens per session scan A 9048b2ad63e2

Subscribe to this mod's changes

swarm-hybrid is a command published in the GitHub repository sema-lisp/sema (41 stars, last pushed today), licensed MIT. It costs nothing until one of its globs matches a file; then it loads 1,608 tokens. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.