run-history-skill-upgrader

1# Run History Skill Upgrader
2 
3## Language Policy
4 
5Write all user-facing output in the user's language. Default to Chinese when the language is unknown.
6 
7## Role
8 
9Turn real run feedback into structural net improvement for an existing skill. Upgrades may add, modify, merge, delete, deprecate, or decide not to change anything.
10 
11## Portability
12 
13This skill is agent-agnostic. It should work in Codex, Claude Code, OpenCode, OpenClaw, Hermes, and similar local agent hosts that can read `SKILL.md` plus optional `references/`, `scripts/`, `evals/`, and `agents/`.
14 
15- Resolve `<skill-dir>` from the directory that contains this `SKILL.md`.
16- Let `<python>` mean the host's available Python launcher: `python3`, `python`, or `py -3`.
17- Do not assume a fixed skill root, shell, home-directory layout, or path separator.
18- Placeholder paths such as `<skill-dir>/scripts/...` describe path segments, not a required separator style. On Windows, use the separator style that your shell or harness accepts.
19- Treat platform-specific helper tooling as optional. Use it only when it is actually available.
20 
21## Mandatory Two-Stage Approval
22 
23This skill always starts in `plan_only`.
24 
251. `plan_only`: read the target skill and the agreed evidence scope, then produce a concrete upgrade plan and stop.
262. `apply_after_approval`: modify files only after the user explicitly approves that specific plan.
27 
28The following are **not** approval by themselves:
29 
30- "upgrade it";
31- "directly edit it";
32- "don't ask me";
33…