Index of every page on this site: llms.txt
All skills
cat-xierluo avatar

/self-evolve

@5fb69a0
Run it yourself

扫描本机 Claude Code、Codex、WorkBuddy、CodeBuddy、ZCode 等 Agent 会话,保留初始化和增量处理记录,低成本初筛后由当前 Agent 提炼经验、迭代与 Skill 沉淀建议,生成可独立打开的 HTML 报告。也用于复盘当前对话、接续每周沉淀或检查扫描覆盖。无需单独配置模型 API Key;不自动应用建议或安装 Skill。

Use this Skill: https://skilld.dev/gh/cat-xierluo/legal-skills/self-evolve

Nothing lands on disk. Nothing to clean up.

Fork this Skill

Edit a local copy. It keeps the author and licence.

references08-evaluation-loop.md

≈672 tokens on demand. Your agent reads this file only when SKILL.md points to it.

提案评估闭环

状态模型

把机器处置、人工决定、实施记录和验证结果存为不同维度:

machine_disposition
  filtered | candidate | duplicate | covered | recommended

human_decision
  pending | accepted | partially_accepted | rejected | observe

application
  not_applied | applied(commit)

outcome
  pending | verified | failed | inconclusive

不要把这些维度压成一个“状态”字段,否则机器重算会覆盖人工判断。

稳定身份与修订

使用以下稳定输入计算 proposal fingerprint:

  • 目标 Skill;
  • proposal 类型;
  • 排序、去重并归一化后的 change points。

再用目标版本、目标内容哈希、目标解析状态、证据 ID、change points、当前唯一 Inbox 逻辑路径和产物内容哈希计算 revision hash。同一 fingerprint 的新证据应更新机器记录;artifact_paths 只保存当前修订绑定,不跨修订累加。revision 未变化时不得重复创建提案。接受前必须确认目标能在当前 inventory 中唯一解析且存在有效内容哈希,通过文件描述符重新读取当前产物核验 SHA-256,并把已核验字节发布为 revision-bound archive 快照。

新颖性判断

生成 recommended 前至少检查:

  1. 目标 Skill 是否仍存在;
  2. 当前版本是否已经覆盖建议;
  3. 同 fingerprint 是否已有终态人工决定;
  4. 同 revision 是否已经 applied/verified;
  5. change points 是否具有跨 session 证据;
  6. 产物是否通过隐私门控。

命中已有人工终态或验证结果时,机器可标记 duplicate 或 covered,不能新建“已接受”记录。

追加审计

人工决定、应用与验证都写入 append-only 事件流。审计日志只保留必要元数据,不复制正文或敏感 before/after 内容。

约束:

  • 接受和部分接受需要通过隐私检查的当前唯一 artifact,在决定时重新核验 hash,并把决定证据固化为 no-clobber archive 快照;
  • 部分接受必须列出具体条目;
  • 拒绝与部分接受必须记录理由;
  • applied 需要 commit hash,验证决定快照,并绑定当前 proposal revision;
  • verified 只能发生在 applied 之后,提交时必须同时给出当前 revision 与被验证的 applied commit;
  • 同一修订的终态决定和验证结果不可静默覆盖。

指标

分别统计:

  • machine disposition 分布;
  • human decision 分布;
  • applied 数量;
  • verified/failed/inconclusive 数量;
  • 从 proposal 到决定、应用、验证的耗时。

不要用机器推荐数冒充接受率,也不要用文件生成数冒充有效迭代数。

Source: SKILL.md on GitHub

skilld matched fixed text patterns in SKILL.md and file names. Patterns miss obfuscated code.

skilld run checks every file with the same patterns. It asks for approval before it loads a Skill with a behavior marked Needs approval.

1 warning18d3 checks · Risk SAFE
  • Gen Agent Trust Hub18d

    The skill is a tool for analyzing local AI agent session history to suggest improvements. It naturally requires access to local logs and performs local script execution, but it includes robust privacy gating and instructions to mitigate risks from untrusted data.

  • Socket18d

    No alerts

  • Snyk18d

    Risk: MEDIUM · 1 issue

Signed by skilld at 5fb69a0. This ties the file your Agent reads to that commit on GitHub. It does not review the instructions.

Last checked against GitHub yesterday.

Activeupdated 4 weeks ago
Other metadata
metadata
{
  "version": "1.28.1",
  "homepage": "https://github.com/cat-xierluo/legal-skills",
  "author": "杨卫薪律师(微信ywxlaw)"
}

README badge

README badge for cat-xierluo/legal-skills/self-evolve