simonwong__skills

skills/simonwong__skills

Fork 0

mirror of https://github.com/simonwong/skills.git synced 2026-09-14 20:54:31 +08:00

Commit Graph

Author	SHA1	Message	Date
Simon	3266e6b77e	chore: make writing skills user-invoked write-article 和 writing-ai-check 的触发词在别的任务里也会出现，自动命中会把顺手的编辑升级成多阶段流程，或直接改写用户没打算交出去评判的文字。两者改为显式调用，rewrite-en2zh 保持 model-invoked。 - SKILL.md 加 disable-model-invocation，版本 0.2.0 → 0.3.0 - 新增 agents/openai.yaml，allow_implicit_invocation: false - README 三张表改为逐个 skill 标注调用方式 - 新增 ADR-0005，ADR-0001 相应指向它 Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>	2026-09-10 20:06:29 +08:00
Simon	d1db2dec4a	feat: add write-article and writing-ai-check skills write-article covers the collaboration flow from idea to finished draft: material inventory with a supplement loop, direction confirmation, then drafting in a fixed order (body, opening, ending, title, layout). Review and AI-flavor check run in fresh subagents. writing-ai-check diagnoses expression in a single draft. Its input and its evidence are the draft itself — it never judges authorship, never outputs an AI rate, and never verifies external facts. Each reference file answers one kind of question; feature lists live only under writing-ai-check/references/. Decisions recorded in docs/adr/. Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>	2026-09-10 18:08:57 +08:00

Author

SHA1

Message

Date

Simon

3266e6b77e

chore: make writing skills user-invoked

write-article 和 writing-ai-check 的触发词在别的任务里也会出现，自动命中会把
顺手的编辑升级成多阶段流程，或直接改写用户没打算交出去评判的文字。两者改为
显式调用，rewrite-en2zh 保持 model-invoked。

- SKILL.md 加 disable-model-invocation，版本 0.2.0 → 0.3.0
- 新增 agents/openai.yaml，allow_implicit_invocation: false
- README 三张表改为逐个 skill 标注调用方式
- 新增 ADR-0005，ADR-0001 相应指向它

Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>

2026-09-10 20:06:29 +08:00

Simon

d1db2dec4a

feat: add write-article and writing-ai-check skills

write-article covers the collaboration flow from idea to finished draft:
material inventory with a supplement loop, direction confirmation, then
drafting in a fixed order (body, opening, ending, title, layout). Review
and AI-flavor check run in fresh subagents.

writing-ai-check diagnoses expression in a single draft. Its input and
its evidence are the draft itself — it never judges authorship, never
outputs an AI rate, and never verifies external facts.

Each reference file answers one kind of question; feature lists live only
under writing-ai-check/references/. Decisions recorded in docs/adr/.

Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>

2026-09-10 18:08:57 +08:00

2 Commits