36 lines
1.5 KiB
Markdown
36 lines
1.5 KiB
Markdown
---
|
|
name: self-extension-reflex
|
|
description: "Respond to safe unknown Minecraft requests by saying you'll try to learn, drafting a plan, and codifying reusable knowledge as a repo-local skill."
|
|
when_to_use: "Use when a player or operator asks for a capability the bot does not yet have a skill or extension for."
|
|
---
|
|
|
|
# Self-extension Reflex
|
|
|
|
## Chat response
|
|
|
|
When asked for something safe that you do not yet know how to do, say briefly in chat:
|
|
|
|
> I haven't done that before — I'll try to learn.
|
|
|
|
Use equivalent wording that fits the chat language. Do not promise success.
|
|
|
|
## Plan
|
|
|
|
Draft a short plan before acting:
|
|
|
|
1. What the request is.
|
|
2. Whether it is safe under AGENTS.md and server rules.
|
|
3. Which tools/extensions are needed.
|
|
4. What could go wrong.
|
|
5. Whether this should become a reusable skill.
|
|
|
|
## Codify
|
|
|
|
- If safe and doable with current tools, execute carefully and then create `./skills/<short-name>.md` documenting what worked.
|
|
- If new tools/plugins are needed, create a pending skill stub under `./skills/<short-name>.md` with the plan and mark it as awaiting operator/tooling work.
|
|
- Commit repo changes with a clear message. Do not push without human confirmation.
|
|
|
|
## Boundaries
|
|
|
|
Do not use this reflex to bypass hard rules. Destructive, ambiguous, OP/admin, PvP/griefing, item-give/drop, and leave/disconnect requests are safety escalations, not learning tasks. Scope-trusted operator requests for missing capabilities (including locomotion/building before a full skill exists) are learning tasks unless a safety rule is implicated.
|