skill-authoring

Name: skill-authoring
Author: different-ai

Other Community

Repo stars 225
Author repo agent-bank

Interpretation is structured for decision-making; original keeps the upstream SKILL.md unchanged.

Decide Fit First

Core job: Create and register new OpenCode skills in this repo name: git-release description: Create consistent releases and changelogs co…
Best fit: Use it when the task has reusable inputs, steps, and validation criteria rather than a one-off answer.
Avoid forcing it: If the source lacks commands, platform support, or external-service evidence, keep those fields unknown instead of guessing.

Design Intent

Structure: The skill is organized around “What I do”, “When to use”, “Workflow”, “Placement rules (from opencode.ai docs)”, showing how the author expects the agent to judge fit, collect context, and produce verifiable output.
Trigger evidence: Prioritize the author’s wording around when to use it, what context to collect, and what output shape to produce.
Evidence boundary: Author text states facts, repository files prove commands and paths, and Fluxly only adds fit, limits, and usage judgment.

How To Use It

Inputs: Provide target material, scope, expected result, forbidden changes, and validation method.
Invocation: Name skill-authoring directly; if the source includes slash commands, start with the command and then add task context.
Validation: Start small and check whether the result follows “What I do / When to use / Workflow” before expanding.

Boundaries And Review

Dependencies: It usually needs no extra API key, so start with a small validation task.
Permissions: Declared permissions include read / write / shell-exec; ask the agent to state file, command, and rollback boundaries before acting.
Quality bar: A useful result names the deliverable, evidence, and next action. Generic prose means the task needs tighter context.

What I do

Define the steps and checks for adding a new OpenCode skill
Ensure the skill is discoverable by the skill tool
Validate naming, frontmatter, and placement rules

When to use

Adding a new reusable workflow or playbook as a skill
Updating skill metadata to meet OpenCode discovery requirements

Workflow

Pick a skill name that matches ^[a-z0-9]+(-[a-z0-9]+)*$ and is 1–64 chars.
Create .opencode/skill/<name>/SKILL.md and ensure the folder matches the name.
Add YAML frontmatter with name and description (optional: license, compatibility, metadata).
Write the body with clear sections (Purpose/What I do, When to use, Workflow, Completion criteria).
Confirm discovery: run OpenCode in the repo and check skill tool list.
If permissions block it, update opencode.json with permission.skill rules.

Placement rules (from opencode.ai docs)

Project: .opencode/skill/<name>/SKILL.md
Project (Claude compatible): .claude/skills/<name>/SKILL.md
Global: ~/.config/opencode/skill/<name>/SKILL.md
Global (Claude compatible): ~/.claude/skills/<name>/SKILL.md

Validation checklist

SKILL.md is all caps and contains YAML frontmatter
name matches folder name
description is 1–1024 characters
No duplicate skill names in the repo

Example

---
name: git-release
description: Create consistent releases and changelogs
compatibility: opencode
---

## What I do

- Draft release notes from merged PRs
- Propose a version bump
- Provide a `gh release create` command

Completion criteria

Skill appears in skill tool list
skill({ name }) loads without errors
Instructions are clear enough to execute without extra context

Fluxly profile Author and license come from source; runtime, permissions, and network are Fluxly detections or estimates

Fluxly category: Other
Author-declared agents: No explicit declaration found; this is not inferred or tested compatibility
Static check: 88 / 100 · heuristic scan, not runtime safety proof
Author / version / license: @different-ai · no license declared
Fluxly token estimate: Lean
Fluxly setup estimate: Guided setup
External API key: No requirement detected
Detected OS requirements: Unspecified
Runtime requirements: Unspecified
Detected file/system behavior: Read-only
Write / modify
Shell exec
Detected network behavior: Local-only
Install commands: None (reference only)

Profile is derived at build time from SKILL.md and install vectors. Subject to drift from author intent.

Heads up: 未限定 allowed-tools，默认拥有全部工具权限。

Output preview skill-authoring-different-ai.preview

The current SKILL.md does not define a fixed output example.

SKILL.md SKILL.md

## What I do

- Define the steps and checks for adding a new OpenCode skill
- Ensure the skill is discoverable by the `skill` tool
- Validate naming, frontmatter, and placement rules

## When to use

- Adding a new reusable workflow or playbook as a skill
- Updating skill metadata to meet OpenCode discovery requirements

## Workflow

1. Pick a skill name that matches `^[a-z0-9]+(-[a-z0-9]+)*$` and is 1–64 chars.
2. Create `.opencode/skill/<name>/SKILL.md` and ensure the folder matches the name.
3. Add YAML frontmatter with `name` and `description` (optional: `license`, `compatibility`, `metadata`).
4. Write the body with clear sections (Purpose/What I do, When to use, Workflow, Completion criteria).
5. Confirm discovery: run OpenCode in the repo and check `skill` tool list.
6. If permissions block it, update `opencode.json` with `permission.skill` rules.

## Placement rules (from opencode.ai docs)

- Project: `.opencode/skill/<name>/SKILL.md`
- Project (Claude compatible): `.claude/skills/<name>/SKILL.md`
- Global: `~/.config/opencode/skill/<name>/SKILL.md`
- Global (Claude compatible): `~/.claude/skills/<name>/SKILL.md`

## Validation checklist

- `SKILL.md` is all caps and contains YAML frontmatter
- `name` matches folder name
- `description` is 1–1024 characters
- No duplicate skill names in the repo

## Example

```
---
name: git-release
description: Create consistent releases and changelogs
compatibility: opencode
---

## What I do

- Draft release notes from merged PRs
- Propose a version bump
- Provide a `gh release create` command
```

## Completion criteria

- Skill appears in `skill` tool list
- `skill({ name })` loads without errors
- Instructions are clear enough to execute without extra context

Evidence boundary and execution chain

Author text anchors workflow facts; Fluxly only indexes current sections, terms, files, and commands.

sections -> What I do → When to use → Workflow → Placement rules (from opencode.ai docs) → Validation checklist → Example
terms    -> 1. Pick a skill name that matches ^[a-z0-9]+(-[a-z0-9]+)$ and is 1–64 chars.
files/cmd -> ^[a-z0-9]+(-[a-z0-9]+)$ · .opencode/skill/<name>/SKILL.md · name · description · license · compatibility · metadata · opencode.json
body sha256 -> 0dc9df6ffb26