write-code-eval

v2026.09.24

Write code evaluators for known failure modes with objective rules. Use when code can check the rule from a trace, with or without a reference answer. Use `write-judge-prompt` when the rule requires interpretation.

GitHub
安装命令
npx skhub add ai-evals-course/write-code-eval
Markdown
SKILL.md

Write a code evaluator

Start with a failure mode found through error analysis. Write one check for that failure mode, much like a unit test that asserts what should hold for each trace.

  1. State the rule and identify the trace fields or reference data the check needs. If the rule requires interpretation, use write-judge-prompt.
  2. Implement the check in the project's language and eval framework. Return a result and a reason in the format that framework expects.
  3. Test known passes and failures, including borderline cases. Run the check on available traces and inspect mistakes. If the rule uses a proxy for human judgment, compare its results with human labels.

Examples

Failure modePossible check
Invalid output structureParse the output and check required fields
Missing or forbidden textMatch a string or pattern
Citation not in retrieved documentsCompare cited IDs with retrieved IDs
Bad tool callCheck arguments against the tool schema or run the call in a safe test environment
Wrong valueCompare the output with a reference value

Choose the check from the failure rule. For a failure with both objective and interpretive parts, check the objective part with code and use a judge for the rest.

发现
标签

此技能尚未发布标签。

版本
最新版本元数据

版本

v2026.09.24

发布时间

Sep 24, 2026

分类

未分类

许可证

Apache-2.0

源路径

skills/write-code-eval

默认分支

main

最新提交

80d5f7b

Tree SHA

478916d