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 wpsnote/wpsnote-skills --skill ie-enginegit clone --depth 1 https://github.com/wpsnote/wpsnote-skillsWrote 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/wpsnote/wpsnote-skills/ie-engine)<a href="https://agentmods.dev/skills/wpsnote/wpsnote-skills/ie-engine"><img src="https://agentmods.dev/badge/skills/wpsnote/wpsnote-skills/ie-engine/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/wpsnote/wpsnote-skills/ie-engine"><img src="https://agentmods.dev/badge/skills/wpsnote/wpsnote-skills/ie-engine.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 | $0.00212 | $0.01899 |
| Opus 5.5 | $0.00085 | $0.00760 |
| Sonnet 5.5 | $0.00042 | $0.00380 |
| Haiku 4.5 | $0.00021 | $0.00190 |
Grade A, and why
ie-engine 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 1mo 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 — 206 lines — stays where its author put it; the contents beside it link to each section on GitHub.
灵感引擎：编排与注意力管理
本技能是灵感引擎（Inspiration Engine）的统一入口和编排中心。 它负责串联三个分层技能，运行完整的灵感生成流水线，并管理何时、如何向用户呈现洞见。
适用场景
被动触发（用户明确请求）
- 用户说"帮我看看过去的笔记里有什么启发"
- 用户说"connect the dots"或"帮我把想法串起来"
- 用户正在做创造性工作，想要从历史笔记中获得灵感
- 用户想全面审视自己某个领域的思考演化
主动触发（Agent 判断用户可能受益）
当你观察到以下信号时，应主动考虑运行灵感引擎并提醒用户：
- 用户正在深入探讨某个主题，且该主题可能在历史笔记中有相关内容
- 用户提出了一个想法或假设，历史笔记中可能有支撑或矛盾的证据
- 用户在不同对话中反复提及类似的概念，暗示这是一个长期关注的方向
- 用户正在做决策或规划，历史经验可能提供有价值的参考
主动触发时，不要直接运行完整流水线，而是先用一句话轻量提示用户，由用户决定是否深入。详见下方「主动触发协议」。
不适用场景
如果用户只需要单一能力，应引导到对应的分层技能：
- 只需要检索笔记 --> ie-retrieve-memory
- 只需要分析笔记关系 --> ie-connect-dots
- 只需要生成洞见 --> ie-generate-insight
- 创建或编辑笔记 --> 直接使用 wpsnote 工具
- 通用问答 --> 直接回答
流水线流程
完整的灵感生成流水线包含三个阶段，按顺序执行：
阶段一：记忆检索（ie-retrieve-memory）
输入：用户的当前上下文/主题/关键词
输出：相关历史笔记列表
↓
阶段二：想法连接（ie-connect-dots）
输入：阶段一的检索结果
输出：聚类、连接假设、长期模式
↓
阶段三：洞见生成（ie-generate-insight）
输入：阶段二的推理结果
输出：洞见、探索建议、演化路径
详细的流水线定义和参数配置请阅读 references/pipeline.md。
编排执行步骤
第一步：理解上下文
在启动流水线之前，先明确用户的意图和当前上下文：
- 确定用户关心的主题或领域
- 确定用户的期望（全面扫描 vs 聚焦某个方向）
- 如果信息不足，向用户询问补充信息
第二步：运行记忆检索
按照 ie-retrieve-memory 的指引，从用户笔记中检索相关历史知识。
参数控制：
- 全面扫描模式：检索 10-15 条笔记，使用多种检索策略
- 聚焦模式：检索 5-8 条笔记，集中在特定主题
第三步：运行想法连接
将检索结果传递给 ie-connect-dots，按照其指引进行推理分析。
参数控制：
- 默认启用全部三种能力（聚类、连接、模式识别）
- 如果检索结果不足 3 条，跳过聚类，只做连接假设
第四步：运行洞见生成
将推理结果传递给 ie-generate-insight，按照其指引生成灵感内容。
参数控制：
- 默认生成 2-3 个洞见 + 2-3 个探索建议
- 如果有明显的时间演化线索，额外生成想法演化路径
第五步：注意力管理与呈现
在向用户呈现结果之前，进行注意力管理判断。
主动触发协议
当 Agent 主动判断用户可能从历史笔记中受益时，遵循以下协议：
第一步：快速预检
不要立即运行完整流水线。先用 ie-retrieve-memory 做一次轻量检索（2-3 条笔记），快速判断是否有相关内容。
第二步：轻量提示
如果预检发现了相关笔记，用一句话提醒用户，格式如下：
[灵感提示] 你之前在「[笔记标题]」等 [N] 条笔记中讨论过相关的想法，要我帮你深入看看有什么启发吗？
关键原则：
- 只用一句话，不展开分析
- 提及具体的笔记标题，让用户判断是否相关
- 以问句结尾，把决定权交给用户
- 不打断用户正在进行的工作流
第三步：等待用户响应
- 用户说「好的」「看看」等肯定回复 --> 运行完整流水线
- 用户忽略或说「先不用」 --> 不再追问，继续当前工作
- 用户说「稍后提醒我」 --> 记住这个话题，在合适的时机再次轻量提示
主动触发的克制原则
- 同一个对话中，主动提示最多出现 2 次，避免干扰用户
- 如果用户已经拒绝过一次主动提示，降低后续触发的敏感度
- 当用户明显处于专注执行状态（连续发出具体指令）时，不要插入提示
注意力管理
What ships with it
1 file beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
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.
- 1mo ago First seen · 206 lines · 212 tokens per session scan A cbd783ad5164
ie-engine is a skill published in the GitHub repository wpsnote/wpsnote-skills (183 stars, last pushed yesterday), licensed MIT. It adds 212 tokens to every session and 1,899 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.
Other skills, from other repositories
causal-memory
Causal memory for agents — install/setup the causal-memory MCP server, then record decisions/outcomes and recall them before acting. Trigger when the user asks to install or set up causal-memory/agent memory, when causal-memory MCP tools are available and the agent faces a non-trivial decision (architecture, debugging…
remem
Use when the user asks Codex to recall prior project context, save durable decisions or bug fixes, inspect remem memory health, or activate remem automatic memory hooks from the Codex plugin.
evo-memory-review
Review and curate an evo-subagent role's prefercmd.md and memory.md using evidence from actual runs. Use when evolution entries are stale, duplicated, contradictory, or too noisy.
ai-memory-coding-conventions
Project coding conventions extracted from chat history. Load when writing new code, naming things, designing APIs, or reviewing PRs. Topics: Cursor pagination is mandatory for all paged GraphQL endpoints.
ai-memory-decision-log
Technical decisions made for this project, with reasoning and rejected alternatives. Load when proposing architectural changes, evaluating alternatives, or asked why a choice was made. Topics: Adopt OAuth 2.0 authorization code flow with PKCE for the customer portal SPA.
ai-memory-system-architecture
System architecture facts: components, data flow, integration boundaries. Load when implementing cross-component features, debugging integration issues, or onboarding to the codebase. Topics: Event sourcing for the billing audit log.