BD ai-llm-evaluation
LLM 应用质量评测与回归测试实操手册——从"感觉不错"到"可度量可门禁"：评测全景与指标体系（正确性/相关性/忠实度/幻觉率/鲁棒性/效率）、评测集构建（黄金数据集/对抗样本/领域评测集/规模估算）、RAG 系统评测（RAGAS 四指标：忠实度/答案相关性/上下文精度/上下文召回）、幻觉检测与度量（事实性幻觉/提示幻觉/上下文矛盾分类与检测方法）、Prompt 回归测试（用例管理/回归门禁/漂移检测/版本对比）、模型对比选型（评测矩阵/成本质量权衡/多模型 A-B/上线决策）、评测流水线与报告（自动化评测/评分聚合/报告模板/上线门禁）。附零依赖本地工具一键查指标、建评测集清单、看 RAG 指标、出对比矩阵、生成评测报告模板。面向 AI 工程师、测试、产品与质量负责人——与 AI 安全红队测试（测安全）互补，本技能测质量。
LLM 应用质量评测与回归测试实操手册——从"感觉不错"到"可度量可门禁"：评测全景与指标体系（正确性/相关性/忠实度/幻觉率/鲁棒性/效率）、评测集构建（黄金数据集/对抗样本/领域评测集/规模估算）、RAG 系统评测（RAGAS…
As a process D 46/100 · Unfinished process — weak spots: result and completion, when it triggers, inputs and preconditions
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.
- 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: 18. Evidence is masked. Grey chips explain why severity was lowered.
Against the Agent Skills spec
- warning
description-long-hermesdescription is 367 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 - note
frontmatter-keyunknown frontmatter key "slug" - note
frontmatter-keyunknown frontmatter key "display_name" - note
frontmatter-keyunknown frontmatter key "displayName" - note
frontmatter-keyunknown frontmatter key "title" - note
frontmatter-keyunknown frontmatter key "description_en"
Process rating: all ten parameters 46/100
- 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
- 60Tools and files. Uses tools (python) that frontmatter does not declare
- 100Steps. 16 steps
- 100Consistency. Name and required fields are in place
- 100Execution cost. Instruction body is 549 tokens
- 100Running it twice. No mutating operations
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
- +5Description has no quoted example phrases that should trigger the skill
- +4Description does not say when NOT to use the skill (false activations)
- +3Output format is not stated: the model decides each time
- +2Single-language instructions
- +3Description length 367: enough signal without eating the budget
- +4Structure: 10 headings
- +3Step-by-step instructions: 16 items
- +4Has examples (1 code blocks)
- +4Reference files are cited in the instructions (8 of 8)
- +1License stated
Quality base 70; lint remarks subtract, signals add up to 100. Result: 68.