AD payroll-data-audit
工资数据审核系统，基于确定性规则引擎 + Python 脚本执行。 全量对齐《工资审核标准流程 SOP》6 步流程。v7.2 修复P0阻断bug：_get_pay_month() 支持 中文格式（"2026年4月"/"2026年04月"/"202604"），RL-003/RL-007 排除逻辑完全生效。 v7.1 修复5个P0阻断bug：全角/半角括号统一、RL-003状态列检查、RL-007新员工排除+上月1号调薪豁免、FR-003实习生排除、YL-006上月1号豁免。 v6.2 新增总审核报告（Master Report）、 规则判定过程详解（judgment）、黄线排除逻辑完善。 v6.2.1 新增编排指南：单节点原则。 Use when user asks to 工资数据审核、薪资校验、算薪逻辑验证、薪酬合规检查、 工资单审核、月度薪资校验、发薪前数据检查、payroll audit、salary check、 wage verification、payroll compliance. 不适用于非薪酬类数据审核、纯算薪操作（非审核）、外部薪酬调研、个税/社保计算. 此技能需手动触发.
As a process D 43/100 · Unfinished process — weak spots: result and completion, when it triggers, inputs and preconditions
How to improve
- 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 · 0
✓ No critical or high findings
Files scanned: 54. Evidence is masked. Grey chips explain why severity was lowered.
Against the Agent Skills spec
✓ No remarks against the Agent Skills spec
Process rating: all ten parameters 43/100
- 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
- 60Tools and files. Uses tools (python) that frontmatter does not declare
- 100Steps. 60 steps
- 100Consistency. Name and required fields are in place
- 100Execution cost. Instruction body is 3025 tokens
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
- -216 emoji in the instructions: noise for the model
- +1No license
- +2Single-language instructions
- +5Description quotes 3 example trigger phrases
- +3Description length 503: enough signal without eating the budget
- +4Structure: 37 headings
- +3Step-by-step instructions: 60 items
- +4Has examples (14 code blocks)
- +3All 13 scripts are documented
Quality base 70; lint remarks subtract, signals add up to 100. Result: 90.