askit-build-output-style
Creates and improves Claude Code output styles (Claude-only response-mode definitions) to the Advanced Skill Library Standard. Use when you need to author a custom output style for Claude Code, define a response mode, or improve an existing output style.
- License 许可
- Apache-2.0
- 最近更新
- 2026-11-14
- Stars
- ★ 3
- Forks
- 0
01 前言 Front Matter
- name
- askit-build-output-style
- metadata
- { "tier": "universal", "version": "0.1.0", "audience": "advanced" }
- description
- Creates and improves Claude Code output styles (Claude-only response-mode definitions) to the Advanced Skill Library Standard. Use when you need to author a custom output style for Claude Code, define a response mode, or improve an existing output style.
02 资源文件 3
- 参考文档examples/golden-1-terse-analysis-style.md6.1 KB
- 参考文档README.md624 B
- 参考文档references/authoring-output-styles.md1.1 KB
03 SKILL.md 正文
askit-build-output-style
Purpose
Author a Claude Code output style: a definition that changes how the agent formats and frames its responses (for example an explanatory or terse mode). Output styles are CLAUDE-ONLY - Codex and the wider agentskills.io ecosystem have no equivalent (Standard sec 2.3) - so this skill emits no Codex artifact and the authored component declares agent-targets: [claude]. Two modes: create scaffolds a new output style; improve refines an existing one. Authoring depth is in references/authoring-output-styles.md. Follows the shared builder contract (../../docs/reference/builder-pattern.md).
When to use
When the user asks to create, scaffold, write, or improve a Claude Code output style (a response mode or response-formatting definition).
create mode
- Brief interview: the style's purpose (when Claude should use it), the response shape it enforces (structure, tone, length), and any required elements. Skip if in context.
- Scaffold the output-style markdown (copy
templates/output-style.md): a name, a description of when it applies, and the response-formatting instructions. - Declare
agent-targets: [claude](Claude-only) and state the asymmetry plainly; do not emit a Codex artifact. - Register it in
library.jsoncomponents.outputStylesas{ name, version, tier, status }. - Evaluate (
node scripts/evaluate.mjs . --json).
improve mode
- Run evaluate and read findings.
- Tighten the description (when-to-use) and the formatting instructions; remove ambiguity about when the style applies and what it requires.
- Re-run to confirm.
Scope
Claude-only (sec 2.3). The build SKILL is portable, but the artifact it produces runs only on Claude Code; state this clearly rather than silently emitting nothing for other targets (the F-06 expectation-setting principle). Output styles are an Advanced, agent-specific capability.
04 同仓库更多 Skills
fx-demo
A fixture skill that exists so the re-derived manifest has a component list to compare. Use only from tests/unit/self-consistency.test.mjs.
fx-demo
A fixture skill that exists so the re-derived manifest has a component list to compare. Use only from tests/unit/self-consistency.test.mjs.
fx-demo
A fixture skill that exists so the re-derived manifest has a component list to compare. Use only from tests/unit/self-consistency.test.mjs.
sf-caller
Calls the worker subagent to do a thing. Use when delegating work to sf-worker.
sf-do-thing
Converts CSV input into a formatted summary. Use when the user asks to summarize or tabulate spreadsheet data.
root-skill-plugin
Summarizes a changelog section into one release-note sentence. Use when the user asks to condense a list of changelog bullets into a single announcement line.