verify

Agent Skills › terrense/LilBot-agent › verify

verify

GitHub

用于在执行前验证实现工作是否正确。根据用户描述或参数生成状态（通过/失败/部分）、证据、风险和下一步建议，强调具体验证而非口头保证。

lilbot/skills/bundled/verify/SKILL.md terrense/LilBot-agent

触发场景

需要验证代码或功能实现正确性时提交完成报告前的最终检查请求

安装

npx skills add terrense/LilBot-agent --skill verify -g -y

更多选项

非标准路径

npx skills add https://github.com/terrense/LilBot-agent/tree/main/lilbot/skills/bundled/verify -g -y

不安装直接使用

npx skills use terrense/LilBot-agent@verify

指定 Agent (Claude Code)

npx skills add terrense/LilBot-agent --skill verify -a claude-code -g -y

安装 repo 全部 skill

npx skills add terrense/LilBot-agent --all -g -y

预览 repo 内 skill

npx skills add terrense/LilBot-agent --list

SKILL.md

Frontmatter

{
    "name": "verify",
    "agent": "verifier",
    "context": "fork",
    "description": "Verify that implementation work is correct before reporting done.",
    "when_to_use": "Use after non-trivial implementation, backend\/API changes, infrastructure changes, or before claiming completion.",
    "allowed-tools": "read_file, list_dir, grep_files, git_status, git_diff, run_tests, agent_open, agent_eval"
}

Verify the work described by the user or by these arguments:

Produce a compact verdict:

STATUS: PASS, FAIL, or PARTIAL. EVIDENCE: commands run, files inspected, and exact outcomes. RISKS: unresolved edge cases or missing coverage. NEXT: smallest next action if not PASS.

Prefer concrete validation over reassurance. If tests cannot be run, explain why and substitute targeted code inspection with file references.