Harmix/open-stanley / open-stanley/skills/learn
learn
VerifiedTurn the user's edits, skips, and results into durable rules — update voice.md and learn/rules.md from draft→final diffs, retune scout scoring from skip reasons, and (only if the user opts in) prepare an anonymized, text-free contribution of rule statistics for the shared Open Stanley skill. Use when the user says "learn from my edits", "here's what I actually posted", "stop doing X", "remember that I never say Y", "why do you keep doing this", or "contribute my improvements".
| File | Kind | Size |
|---|---|---|
| SKILL.md | skill md | 4.4 KB |
| references/contribution-schema.md | markdown | 1.4 KB |
references/contribution-schema.md
View on GitHub# Contribution schema (v1)
Everything is a count or a range. There is deliberately no field that can hold free text longer than a rule id.
```json
{
"schema": 1,
"plugin_version": "0.1.0",
"month": "2026-09",
"salt_rotated": true,
"platforms": {"linkedin": true, "x": true},
"volume": {"posts": "1-5", "comments": "10-30", "scans": "20-60"},
"audit": {
"fired": {"lexicon": 14, "not-x-but-y": 3, "choppy": 6, "closing-question": 2},
"user_reverted_fix": {"closing-question": 2}
},
"edits": {
"cut-setup-sentence": 9, "shortened-25-50pct": 5, "replaced-word": 7,
"changed-ending-to-question": 4, "removed-number": 2
},
"word_swaps": [["innate", "*"], ["leverage", "*"]],
"skip_reasons": {"too-abstract": 6, "off-lane": 4, "already-commented": 2},
"new_rule_candidates": ["ban:agreement-opener-comments"],
"runtime": {"host": "claude-code|cowork|codex|openclaw|other", "browser": true, "capabilities": ["meetings", "messages", "browser"]}
}
```
Rules for `word_swaps`: only the *draft* word is kept (it is by definition a generated word); the replacement is always `*`. Rule ids come from the linter's fixed set or from `new_rule_candidates` as `ban:<slug>` / `allow:<slug>`. Ranges for volume are fixed buckets. Capabilities are category names, never server names. Maintainers drop any bucket with fewer than 5 contributors before publishing aggregates.
Related skills
| Skill | Stars | Updated | Works with | Status | Install |
|---|---|---|---|---|---|
| reverse-engineer-anythingmorluto/reaReverse engineer native, managed, Electron/JavaScript, packaged, firmware, and browser targets with REA, and build agent environments (AgentEnv bundles) of web applications from a live instance or a capture, and of Linux desktop apps with a task from a request. Use shipped-artifact or requested runtime evidence to explain features, compare versions, decompile code, or guide a reconstruction. Skip REA for ordinary source-repository architecture analysis. | Stars: 471k | Updated | Claude CodeCodex | Verified | Install |
| diagnosing-superpowersobra/superpowersUse when a superpowers session went wrong and your human partner wants to know why — repeated work, ignored plans, stumbles, poor results, a skill that didn't fire, "it took too long", "why is it so expensive", "what is it doing" — or wants to build a bug report for the superpowers maintainers, for the current session or a past one identified by id or path, on any harness. | Stars: 322k | Updated | Claude CodeCodex | Verified | Install |
| dispatching-parallel-agentsobra/superpowersUse when facing 2+ independent tasks that can be worked on without shared state or sequential dependencies | Stars: 322k | Updated | Claude CodeCodex | Verified | Install |
| executing-plansobra/superpowersUse when executing an implementation plan in the current session as the implementer yourself — your human partner chose inline execution, or no subagent tool is available | Stars: 322k | Updated | Claude CodeCodex | Verified | Install |