y3owk1n/neru / .agents/skills/neru-add-platform-feature
neru-add-platform-feature
VerifiedImplement or stub Neru functionality for a specific OS/backend: port contract, build-tagged file slots, factory wiring, CodeNotSupported stubs, capability matrix, contract tests, and the docs ownership table. Use for any change under internal/adapter/platform or any *_darwin/_linux/_windows file. Not for pure-Go domain logic.
| File | Kind | Size |
|---|---|---|
| SKILL.md | skill md | 2.8 KB |
| agents/openai.yaml | config | 309 B |
SKILL.md
View on GitHub---
name: neru-add-platform-feature
description: "Implement or stub Neru functionality for a specific OS/backend: port contract, build-tagged file slots, factory wiring, CodeNotSupported stubs, capability matrix, contract tests, and the docs ownership table. Use for any change under internal/adapter/platform or any *_darwin/_linux/_windows file. Not for pure-Go domain logic."
---
# Platform work in Neru
Platform code carries Neru's strictest guardrails, and this is where agents
most often guess. A change that lands cleanly sits in an existing file slot,
crosses the boundary through a port, returns `CodeNotSupported` wherever it is
unimplemented, and reports itself honestly in the capability matrix.
## Before writing code
1. Read `internal/adapter/platform/AGENTS.md`, the single home for the rules
this skill obeys: the One Rule, file slots, the factory and Linux's runtime
compositor axis, loud stubs, capability honesty, coordinates, generated
Wayland bindings. Add the `darwin/` or `linux/` guide for the native
boundary you touch.
2. Read `internal/adapter/platform/profile.go`. It is the source of truth for
each subsystem's backend family, primary-modifier expectations, and whether
a backend needs CGO. CGO is a **per-backend** decision, not a per-OS one.
3. Read the port you're implementing in `internal/ports/`. If the capability
doesn't have a port yet, define the interface there first and add a mock in
`internal/ports/mocks`.
## Tests
- Unit tests with port mocks stay platform-neutral.
- Real-OS behavior goes in `*_integration_<os>_test.go` tagged
`//go:build integration && <os>`.
- Contract tests pin `CodeNotSupported` per subsystem, not per stub. Update the
subsystem's existing one if it has one, and write a new one when a caller
could read the stub's `nil` as success, so a later "implementation" that
silently no-ops fails a test. `internal/adapter/platform/AGENTS.md` names the
ones that exist.
## Docs (same change, not a follow-up)
`docs/contributing/porting.md` has the ownership table ("Documentation Checklist"),
and each fact has exactly one home. Capability status goes in
`docs/reference/platform-support.md`, the API or protocol behind it in
`docs/contributing/platform-internals.md`, and never either in
`docs/contributing/architecture.md` (shape only).
Linux setup specifics go in `docs/guide/linux.md` / `docs/guide/linux-desktops.md`.
## Verify
```bash
just lint # depguard enforces the One Rule
go test ./internal/architecture/
just test-foundation
```
Cross-compile the platforms you touched (`just build-linux`,
`just build-windows`) even when you can't run them, because build tags hide breakage
from a darwin-only `just build`. Then the standard pre-commit gate, and ask a
maintainer to run integration tests on the target OS if you can't.
Related skills
| Skill | Stars | Updated | Works with | Status | Install |
|---|---|---|---|---|---|
| finishing-a-development-branchobra/superpowersUse when implementation is complete, all tests pass, and you need to decide how to integrate the work | Stars: 322k | Updated | Claude CodeCodex | Verified | Install |
| systematic-debuggingobra/superpowersUse when encountering any bug, test failure, or unexpected behavior, before proposing fixes | Stars: 322k | Updated | Claude CodeCodex | Verified | Install |
| test-driven-developmentobra/superpowersUse when implementing any feature or bugfix, before writing implementation code | Stars: 322k | Updated | Claude CodeCodex | Verified | Install |
| skill-creatoranthropics/skillsCreate new skills, modify and improve existing skills, and measure skill performance. Use when users want to create a skill from scratch, edit, or optimize an existing skill, run evals to test a skill, benchmark skill performance with variance analysis, or optimize a skill's description for better triggering accuracy. | Stars: 198k | Updated | Claude CodeCodex | Verified | Install |