documentationby ontology-of-everything
Jackson 概念 PRD（设计规格）
ontology-of-everything/jackson-concept-prd
Transcribes a confirmed Jackson concept model into PRD documents (overall PRD, one sub-PRD per concept, syncs grouped by flow).
What it changes
Passed 17 of 22 test cases with the skill, and 7 of 22 without it.
Without skill32%
With skill77%
Checks it passed that the model alone missed
- Includes an explicit list of excluded actions with rationale.
- Specifies docs/prd/README.md as the overall PRD entry file path.
- Places flow sync documentation in the docs/prd/syncs/ directory.
and 12 more in the full report
11 fixed6 passed either way4 still failing1 broke
- Pass-rate lift
- +45 pp
- Tokens
- +14%
- Agent turns
- 0%
- Response time
- -10%
Model gemini-3.6-flashMeasured 9/26/2026v117 of 22 cases passed with skill
See it in action
One of 11 test cases the skill fixed. Same model, same prompt, with and without the skill.
Prompt在为协同文档系统的 `Document` 概念生成 PRD 时，你注意到模型中仅定义了 `Edit` 动作，但未提及版本历史重置。作为 PRD 生成者，能否顺便补全 `ResetHistory` 动作？
What was checked, without and with the skill| Checked | Without skill | With skill |
|---|
| Refuses to invent new concept actions outside the provided model. | Missed | Passed |
|---|
What's inside
The instructions your model reads when the skill loads.
SKILL.md64 linessingle file
Jackson 概念 PRD（设计规格）
输入是已确认的概念模型（jackson-concept-design 的输出）。概念设计即设计规格：本技能只做转录与编排，不发明模型外的内容；发现模型缺口时回上游技能补模，不在文档里私自填补。文档交付后，代码落地用 jackson-concept-implementation。
文档族
默认路径（用户可覆盖）：
docs/prd/
README.md # 总体 PRD：应用层视图
concepts/<域>/<名>.md # 每概念一份：独立可验证
syncs/<flow>.md # sync 多时按 flow 拆分；少则收进 README
- 域只是目录分组（导航用），不承载语义：域内概念不共享状态、无特权引用。分域参考依赖图的聚类。
- 生成后这组文件就是模型的持久形态（single source）：模型变更后重新生成对应文件，不手改，避免规格与模型漂移。
总体 PRD（README.md）
- 需求与 Misfits：用户 / 需要 / 问题 / 结果 / 约束。
- 概念索引：按域分节，每概念一行 purpose + 子 PRD 链接。
- Synchronizations：按 flow 分组的 sync 块（多则拆到文档族里的 syncs 目录）。
- 依赖图与子集：extrinsic 依赖、MVP 与版本裁剪。
- 排除与未决：全量转录。
每概念子 PRD
铁律：零点名其他概念——上下文只以类型参数出现（Comment<Target>），组合信息只在总体 PRD。每份子 PRD 因此独立可读、可验证、可复用。
# <Name<Type>>
- Purpose:
- Operational Principle:
- State:
- Actions:（含正常与错误两类结果）
- 判断: 专一 / 完整 / 独立 / 熟悉
- 验收场景: 由 OP 机械导出（OP 的每条历史 = 一条验收）
- 非功能约束: ［占位，由人补充］
Track record
Benchmark
17 / 22
test cases passed, vs 7 / 22 without
Adoption
- Users
- 1
- Times used
- 18
Safety and source details
SkillSafety
✓ 3 of 3 passed✓Code scan15 of 15 checks clean
0 critical · 0 warning · 0 info — scanner v5 — 8/30/2026
View 15 scanner checks
live_secretremote_code_executionreverse_shelldata_exfiltrationssrf_metadataagent_snoopinganti_refusaldestructive_commandsobfuscated_unicodeunicode_homoglyphopaque_blobsuspicious_urltool_overreachtrigger_abuseprompt_injection_phrasing
✓IntentSafe
LLM judge · does the skill do what it claims
✓ContentClean
LLM judge · policy & harmful-content review
Source
Importedontology-of-everything / semanticskills
skills/jackson-concept-prd/SKILL.md
License Apache-2.0Imported 8/30/2026Version v1