Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx agentmods add instructions/eduardozf/wire-lang/agents-mdgit clone --depth 1 https://github.com/eduardozf/wire-langWhat 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.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5 | $0.00549 | $0.00549 |
| Opus 5 | $0.00275 | $0.00275 |
| Sonnet 5 | $0.00110 | $0.00110 |
| Haiku 4.5 | $0.00055 | $0.00055 |
Grade A, and why
wire-lang AGENTS.md 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 2d 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.
How it starts
The opening of the file, as written. The whole thing — 47 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Agent Instructions
This repository includes an Agent Skills-compatible user skill at
skills/wire-lang/SKILL.md.
Use that skill when helping someone write, revise, explain, or troubleshoot
Wire Lang .wire source. The skill is for Wire Lang users and source authoring;
it is not the development guide for implementing the compiler or renderer.
Keep the project documented as you work. When behavior, vocabulary, scope, or architecture changes, update the relevant source-of-truth document in the same change:
docs/CONTEXT.mdfor domain vocabulary and resolved terminology.docs/MVP.mdfor product scope, language semantics, APIs, diagnostics, and implementation contracts.docs/adr/for hard-to-reverse architectural decisions that would be surprising without context.skills/wire-lang/for user-facing authoring instructions, examples, and syntax guidance.
The symbol bench picker is derived from the standard library, so new component
types appear there automatically. When a type needs example props to render
meaningfully (e.g. Header/IC need a pins=[...] list), add a SAMPLE_PROPS
entry in tools/symbol-bench/bench.js in the same change.
Pull Requests
When a PR changes rendered output — layout, symbols, SVG, anything an author would see differently — include before/after images in the PR description. PRs with no visual diff (docs, tooling, refactors, CI, internal APIs) do not need them.
Produce each image with the bundled SVG renderer so it reflects real output, not
a mockup: render the same .wire source on main (before) and on the branch
(after) and place them side by side, main on the left. scripts/wire-to-png.mjs
rasterizes a .wire file to PNG via @resvg/resvg-js and is a good starting
point. Choose small, focused schematics that isolate the change, and prefer a
few cases over one busy diagram (e.g. each direction a hint can move, plus a
no-op case).
Host the images on a throwaway demo/<topic>-assets branch under
docs/assets/<topic>/, pushed separately, and reference them in the PR body by
raw URL (https://raw.githubusercontent.com/<owner>/wire-lang/demo/<topic>-assets/...).
Keeping them off the feature branch keeps binary assets out of the PR's diff and
out of main. Note in the PR that the images live on that branch.
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.
- 2d ago First seen · 47 lines · 549 tokens per session scan A a91e9a9a2301
wire-lang AGENTS.md is an instructions file published in the GitHub repository eduardozf/wire-lang (4 stars, last pushed 8d ago), licensed MIT. It adds 549 tokens to every session, about $0.0027 per session on Opus 5. 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 2026-08-31.
Other instructions, from other repositories
kicad-happy GEMINI.md
Instructions for aklofas/kicad-happy, covering kicad-happy, agent guidelines, skills, running analysis and schematic analysis.
OpenGeometry AGENTS.md
Instructions for OpenGeometry-io/OpenGeometry, covering agents.md, 1. project identity, 2. architecture, runtime flow and build pipeline.
OpenGeometry copilot-instructions.md
Instructions for OpenGeometry-io/OpenGeometry: All agent guidance lives in AGENTS.md. Read that first.
OpenGeometry CLAUDE.md
Instructions for OpenGeometry-io/OpenGeometry: All agent guidance lives in AGENTS.md. Read that first.
kicad-studio-kit AGENTS.md
AGENTS.md instructions for oaslananka/kicad-studio-kit, covering kicad studio kit agent instructions, repository boundaries, required first reads, local commands and mcp safety defaults.
kicad-studio-kit copilot-instructions.md
Copilot instructions for oaslananka/kicad-studio-kit, covering github copilot instructions, project shape, coding rules, validation and mcp defaults.