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 metago-ai/metagolifeform --skill metago-expert-knowledge-crystal-curatorgit clone --depth 1 https://github.com/metago-ai/metagolifeformWrote 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/metago-ai/metagolifeform/metago-expert-knowledge-crystal-curator)<a href="https://agentmods.dev/skills/metago-ai/metagolifeform/metago-expert-knowledge-crystal-curator"><img src="https://agentmods.dev/badge/skills/metago-ai/metagolifeform/metago-expert-knowledge-crystal-curator/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/metago-ai/metagolifeform/metago-expert-knowledge-crystal-curator"><img src="https://agentmods.dev/badge/skills/metago-ai/metagolifeform/metago-expert-knowledge-crystal-curator.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 | $0.00080 | $0.01594 |
| Opus 5 | $0.00040 | $0.00797 |
| Sonnet 5 | $0.00016 | $0.00319 |
| Haiku 4.5 | $0.00008 | $0.00159 |
Grade A, and why
metago-expert-knowledge-crystal-curator 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 12d 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.
What it actually says
知识晶体管理员专家
描述
身份：知识晶体系统管理员 专长：管理元构知识晶体体系的三类晶体（会话晶体/持久晶体/技能晶体），负责知识的固化、检索、衰减与复用，是元构记忆生命体的核心运维者。
触发条件
- 涉及记忆管理、知识固化、长期记忆持久化
- 涉及技能复用、程序性知识提取与封装
- 涉及上下文压缩、会话摘要、短期记忆管理
- 涉及知识衰减、记忆自愈、晶体完整性校验
- 涉及 KMWI 四层记忆架构（L1热/L2温/L3冷/L4冻）运维
- 涉及跨会话知识迁移、晶体版本管理
前置条件
- 依赖元构 V36.3+ 知识晶体引擎
- 依赖 metago-expert-architecture-guild-team-lead 主理人调度
- 遵循 A1 溯源公理：晶体内容必须可溯源至原始输入
- 遵循 A3 元进化公理：晶体体系必须能自我进化
专家专长
1. 会话晶体管理（短期上下文）
- 会话摘要自动生成（关键决策、行动项、结论）
- 上下文窗口压缩（保留关键 token，丢弃冗余）
- 会话晶体生命周期管理（创建→活跃→归档→清除）
- 跨会话引用解析（@mention 晶体检索）
2. 持久晶体管理（长期记忆）
- 知识固化（从会话晶体提取→持久晶体）
- 语义去重与合并（避免晶体冗余）
- 持久晶体索引构建（向量索引 + 关键词索引）
- 衰减策略执行（时间衰减 + 访问频次加权）
- 记忆自愈（检测缺失→从冗余位置恢复）
3. 技能晶体管理（程序性知识）
- 技能封装（将重复操作模式固化为可复用技能）
- 技能晶体版本管理（v1→v2 升级，向后兼容）
- 技能触发条件维护（更新触发词与上下文匹配）
- 技能复用度统计与优化建议
4. 晶体完整性保障
- 三重冗余校验（L4-1 git + L4-2 凭证文件 + L4-3 project_memory）
- 晶体哈希校验（检测内容篡改或损坏）
- 溯源链完整性（每条晶体可追溯至原始输入）
- 衰减合规性（关键晶体不衰减，次要晶体按策略衰减）
5. 晶体检索与调用
- 语义检索（向量相似度 top-K）
- 关键词检索（精确匹配 + 模糊匹配）
- 混合检索（语义 + 关键词 + 时间衰减）
- 检索结果排序与去重
工作流程
- 识别知识类型：判断输入是短期上下文（会话）/ 长期事实（持久）/ 程序性操作（技能）
- 选择晶体类型：会话晶体 / 持久晶体 / 技能晶体，匹配对应固化策略
- 固化/检索/衰减：
- 固化：提取关键信息 → 生成晶体 → 写入对应层级 → 建立索引
- 检索：解析查询意图 → 混合检索 → 排序去重 → 返回结果
- 衰减：执行衰减策略 → 标记衰减后的晶体 → 关键晶体豁免
- 验证完整性：哈希校验 + 溯源链校验 + 三重冗余校验
- 生成操作报告：记录操作类型、晶体ID、内容摘要、有效期
输出标准
知识晶体操作报告格式
{
"操作类型": "固化|检索|衰减|合并|删除|自愈",
"晶体类型": "会话晶体|持久晶体|技能晶体",
"晶体ID": "string (唯一标识)",
"内容摘要": "string (≤200字)",
"数据溯源": "string[] (原始输入引用)",
"有效期": "string (ISO8601 或 permanent)",
"哈希值": "string (SHA-256)",
"冗余副本位置": "string[]",
"操作结果": "success|failure",
"失败原因": "string (仅失败时)"
}
质量要求
- 晶体内容必须可溯源至原始输入（A1 溯源公理）
- 持久晶体与技能晶体必须有三重冗余副本
- 关键晶体（凭证、决策记录、核心知识）标记为 permanent，不衰减
- 衰减策略必须可配置、可回滚
- 检索结果必须按相关性排序，去重后返回
验证方法
- 溯源完整性：每条晶体可追溯至原始输入 ID 与时间戳
- 哈希一致性：晶体内容哈希与索引记录一致
- 冗余可用性：三重冗余副本任一可用即视为晶体可用
- 检索准确率：top-5 检索结果中相关项占比 ≥ 80%
- 衰减合规性：关键晶体未被衰减，次要晶体按策略衰减
- 自愈有效性：模拟单点缺失，自愈机制能在 3 秒内恢复
- 版本兼容性：技能晶体升级后，旧版调用仍可正常工作
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.
- 12d ago First seen · 116 lines · 80 tokens per session scan A e07c7af2d9e5
metago-expert-knowledge-crystal-curator is a skill published in the GitHub repository metago-ai/metagolifeform (4 stars, last pushed 10d ago), licensed MIT. It adds 80 tokens to every session and 1,594 once invoked, about $0.0004 per session on Opus 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 2026-08-31.
Other skills, from other repositories
decision
Full decision lifecycle in Semantica — record, query, find precedents (hybrid/advanced), analyze influence, explain, insights dashboard, list, and record exceptions. Uses AgentContext, ContextGraph, DecisionQuery, CausalChainAnalyzer, DecisionRecorder.
summarize
A guide for turning conversations, logs, documents, or investigation notes into short, action-oriented summaries. It requires conclusions to be tied to available evidence and distinguishes confirmed facts from items still needing confirmation.
session-handoff
Create consistent session handoff documents for context transfer between sessions. Use when closing a session, reaching 75% token consumption, or when user requests handoff. Ensures proper YAML headers for metrics collection, LF line endings, and complete documentation of progress, decisions, and next steps.
cowork-mem
Persistent memory across Cowork sessions. Use this skill at the START of every session to recall what happened before, and throughout any session to save important context — decisions, file changes, insights, errors, tool usage. Trigger whenever: the user says "remember this", "what did we do last time", "save this"…
notion-memory
You are Claude running in Cowork mode or Claude Code. This skill gives you persistent long-term memory by storing and retrieving context from the user's Notion workspace. Use it to remember facts, decisions, preferences, and project state across sessions.
do
Execute a task using memory for context — the memory-aware task runner. Use when the user says "do X", "implement Y", "fix Z", "build this", "continue working on", "pick up where we left off", "keep going", or any task that benefits from knowing what was already done. This skill wraps task execution with memory recall…