Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx skills add L-LesterYu/OpenClaw-hot-skills-zh --skill persona-learninggit clone --depth 1 https://github.com/L-LesterYu/OpenClaw-hot-skills-zhWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/skills/l-lesteryu/openclaw-hot-skills-zh/persona-learning)<a href="https://agentmods.dev/skills/l-lesteryu/openclaw-hot-skills-zh/persona-learning"><img src="https://agentmods.dev/badge/skills/l-lesteryu/openclaw-hot-skills-zh/persona-learning/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/skills/l-lesteryu/openclaw-hot-skills-zh/persona-learning"><img src="https://agentmods.dev/badge/skills/l-lesteryu/openclaw-hot-skills-zh/persona-learning.svg" alt="Reviewed on agentmods" width="80" height="20"></a>What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00208 | $0.03500 |
| Opus 5.5 | $0.00083 | $0.01400 |
| Sonnet 5.5 | $0.00042 | $0.00700 |
| Haiku 4.5 | $0.00021 | $0.00350 |
Grade A, and why
persona-learning scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 2mo ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 473 lines — stays where its author put it; the contents beside it link to each section on GitHub.
人格学习系统
本技能实现 AI 人格操作系统的学习捕获、审查和晋升系统。提供结构化机制来捕获学习内容、错误和功能请求,然后将有价值的模式提升到长期记忆。
阶段 1:捕获
将结构化条目记录到 ~/workspace/.learnings/ 中的相应文件:
步骤 1.1:确定条目类型
分类条目:
- 学习内容 →
LEARNINGS.md(洞察、发现、成功的方法) - 错误 →
ERRORS.md(失误、失败、错误假设) - 功能请求 →
FEATURE_REQUESTS.md(期望的功能、工具缺口、工作流改进)
步骤 1.2:确保学习目录存在
if not exists(~/workspace/.learnings/):
create directory ~/workspace/.learnings/
if not exists(target_file):
create target_file with header
步骤 1.3:格式化条目
使用以下字段结构化条目:
## [YYYY-MM-DD] 类别:简要标题
**描述:** 对学习到的内容/遇到的错误/需要的功能的清晰说明
**上下文:** 导致此学习/错误/需求的情况
- 相关项目/任务
- 尝试了什么
- 目标是什么
**解决方案/操作:**
- 学习内容:什么有效、为什么有效、何时应用
- 错误:根本原因、如何修复、预防策略
- 功能请求:提议的解决方案、实现前的替代方案
**标签:** #关键词1 #关键词2 #关键词3
**状态:** 活跃 | 已晋升 | 已解决
步骤 1.4:追加到相应文件
Read target_file
Append formatted_entry to target_file
Write updated target_file
Confirm to user: "条目已捕获到 {target_file}"
阶段 2:审查
扫描学习文件以发现模式和晋升候选。
步骤 2.1:扫描所有学习文件
learnings = Read ~/workspace/.learnings/LEARNINGS.md
errors = Read ~/workspace/.learnings/ERRORS.md
feature_requests = Read ~/workspace/.learnings/FEATURE_REQUESTS.md (if exists)
entries = parse_all_entries(learnings, errors, feature_requests)
步骤 2.2:检测模式
查找以下内容:
- 重复错误(相同错误类型 3 次以上)
- 常见主题(标签出现在 5 个以上条目中)
- 晋升候选(状态:活跃,出现 3 次以上)
- 已解决模式(多个条目具有相同根本原因)
patterns = {
"repeated_errors": [],
"common_tags": {},
"promotion_candidates": [],
"resolved_themes": []
}
for entry in entries:
if entry.status == "Active":
increment tag_counts[entry.tags]
if entry appears 3+ times (by description similarity):
add to promotion_candidates
if entry.type == "error" and similar_errors >= 3:
add to repeated_errors
步骤 2.3:生成审查摘要
输出格式:
# 学习审查 - [日期]
## 统计摘要
- 学习内容总数:{数量}
- 错误总数:{数量}
- 功能请求:{数量}
- 已晋升条目:{数量}
## 检测到的模式
### 重复错误({数量})
1. [错误主题] - {数量} 次出现
- 最近出现:[日期]
- 预防策略:[如有]
### 常见主题({数量} 个标签出现 5 次以上)
- #标签1:{数量} 个条目
- #标签2:{数量} 个条目
### 晋升候选({数量})
1. [标题] - {数量} 次出现
- 首次出现:[日期]
- 最近出现:[日期]
- 准备晋升到 MEMORY.md
### 已解决模式
1. [主题] - {数量} 个实例已解决
- 解决方案:[摘要]
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 2mo ago First seen · 473 lines · 208 tokens per session scan A e3c144481f39
persona-learning is a skill published in the GitHub repository L-LesterYu/OpenClaw-hot-skills-zh (57 stars, last pushed 7mo ago), licensed MIT. It adds 208 tokens to every session and 3,500 once invoked, about $0.0008 per session on Opus 5.5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.
persona-learning compared with skills from other repositories
| Mod | Type | Stars | Last push | Licence | Scan | Tokens | Works with |
|---|---|---|---|---|---|---|---|
| persona-learningL-LesterYu/OpenClaw-hot-skills-zh | Skill | 57 | 7mo ago | MIT | A | 208 | Claude Code |
| self-improve-with-tinyfishtinyfish-io/tinyfish-cookbook | Skill | 2.3k | 4d ago | MIT | A | 68 | Claude Code, Codex |
| flow-next-auditgmickel/flow-next | Skill | 718 | yesterday | MIT | A | 46 | Claude Code, Codex |
| learn-from-fixoliver-kriska/claude-elixir-phoenix | Skill | 573 | 2d ago | MIT | A | 46 | Claude Code |
| mindosGeminiLight/MindOS | Skill | 704 | 16d ago | MIT | A | 136 | Claude Code |
| openclaw-knowledge-coachSibo-Zhao/OpenPraxis | Skill | 415 | 17d ago | MIT | A | 65 | Codex |
| lore-lessonshsusul/lore | Skill | 134 | yesterday | Apache-2.0 | A | 44 | Claude Code |