DF zhihu-content-pack
知乎发文内容包全流程：选题→强制全面核实（先核实后写稿；**核实范围按交付清单逐件枚举**，含标题/图注/图内文字/执行单/口径说明段/版本号；抓一手源全文+逐条建表+反向验证"空白"类断言+主体性断言找当事人自述+口径一致性自检）→写稿→生成可粘贴 HTML→出图（有硬性字号/字数下限）→**回答交付标准包（源稿+粘贴版+发布执行单+配图+内嵌图注，回答类默认整套交付）**→发布指引与自查→发布后回填权属台账与 SHA-256 指纹。当用户要求"发知乎/写知乎文章/知乎回答/知乎内容包"时触发。核心铁律：全面核实是默认动作不等用户问；回答一律按标准包交付不等用户提；核实范围不留漏网区；**断言分级标注 + 「本次未验」栏为交付必备（M1/M5），无锚点的"已完成"视为过度声称**；**交付前跑确定性门禁 `delivery_guard.py`（fail-closed，不通过不许交付）**；**事实层走 L2 独立验证（信息隔离投喂 + 抽验其锚点）并过「措辞强度」三件套（溯源链标注/法律前提重申/口径并存披露）**；**L2 的锚点必须用 `anchor_verify.py` 做机器全量校验（手抽不够），锚点须逐字照录不得改写**；**不得把"用了 reasoning 路由"当成"跨族验证"来声称**。
知乎发文内容包全流程：选题→强制全面核实（先核实后写稿；核实范围按交付清单逐件枚举，含标题/图注/图内文字/执行单/口径说明段/版本号；抓一手源全文+逐条建表+反向验证"空白"类断言+主体性断言找当事人自述+口径一致性自检）→写稿→生成可粘贴…
As a process F 29/100 · No process to follow — References files that are not bundled: assets/_cover-test/, assets/_superseded/, assets/_legacy-html/
How to improve
- Say in the description WHEN to use the skill ("use when…", example requests): that is the agent's main cue.
- For Hermes the description must be one sentence under 60 characters; move the conditions to a "When to Use" section.
- The SKILL.md body is over 5,000 tokens: move reference detail into references/ and load it when needed.
- The text references files that are not there: add them or drop the references.
- Your own cases (evals/evals.json, 4–6 real requests with expected answers): the full check would then run those instead of a model-drafted suite.
- A spec.yaml with trigger phrases and assertions — a behaviour contract for CI; `skilltest init` writes a template.
Guard findings · 1
✓ No critical or high findings
Medium and low: 1
✓ Guard found no suspicious behaviour. 1 matches are attack strings quoted in this security skill's own documentation.
Files scanned: 8. Evidence is masked. Grey chips explain why severity was lowered.
Against the Agent Skills spec
- warning
frontmatter-yamlSKILL.md: the frontmatter is not valid YAML (YAML parse error: Nested mappings are not allowed in compact mappings at line 5, column 17: description_en: Full pipeline for publishing to Zhihu (Q&A / columns): topic re… ^ ); fields were read line by line. The usual cause is a colon inside an unquoted value - warning
description-long-hermesdescription is 561 chars; the Hermes authoring standard requires ≤ 60 (one sentence, ending with a period) - warning
description-no-whenneither description nor a "## When to Use" section says when to use the skill - warning
body-longSKILL.md body ≈ 9270 tokens (recommended < 5000); move details to references/ - warning
missing-refreference to a missing file: assets/_cover-test/ - warning
missing-refreference to a missing file: assets/_superseded/ - warning
missing-refreference to a missing file: assets/_legacy-html/ - warning
missing-refreference to a missing file: assets/figN-*.png - note
frontmatter-keyunknown frontmatter key "slug" - note
frontmatter-keyunknown frontmatter key "displayName" - note
frontmatter-keyunknown frontmatter key "description_en" - note
edit-residuethe text marks something as outdated (lines 513): check that old rules are not kept next to new ones — the full check reads the text for contradictions
Process rating: all ten parameters 29/100
- 0Tools and files. 4 referenced file(s) missing: assets/_cover-test/, assets/_superseded/, assets/_legacy-html/
- 0Result and completion. Does not say what the result is
- 0Inputs and preconditions. Does not say what the process needs to start
- 0Failures and branches. Linear process with no failure handling
- 0Progress reporting. Says nothing while it works
- 20When it triggers. No condition that starts the skill
- 30Running it twice. 1 mutating operations with no state check
- 40Execution cost. Instruction body is 9270 tokens: crowds the task out of the window
- 100Steps. 130 steps
- 100Consistency. Name and required fields are in place
- low 13 top-level sections: this looks like several domains in one skill
- low The response is described with custom markup (10 tags): a typed call is more reliable
Everything here is measured from the skill text rather than judged by a model, so the numbers are checkable. A parameter weighs more when it is a more common reason for the process to stall.
Quality signals
- +4Description does not say when NOT to use the skill (false activations)
- +3Output format is not stated: the model decides each time
- -285 emoji in the instructions: noise for the model
- +2Single-language instructions
- +5Description quotes 2 example trigger phrases
- +3Description length 561: enough signal without eating the budget
- +4Structure: 45 headings
- +3Step-by-step instructions: 130 items
- +4Has examples (10 code blocks)
- +1License stated
Quality base 70; lint remarks subtract, signals add up to 100. Result: 35.