返回技能列表

adaptic

pjt222
更新于 2 days ago
7 次查看
17
2
17
在 GitHub 上查看
aidesign

关于

Adpatic is a master skill that orchestrates a 5-step synoptic cycle to achieve panoramic synthesis across three or more domains, producing a unified understanding instead of a sequential compromise. Use it when domain interactions are critical or before major architectural decisions affecting multiple stakeholders. It integrates steps like meditation, observation, and gestalt integration into a coherent process for complex, cross-domain problems.

快速安装

Claude Code

推荐
主要方式
npx skills add pjt222/agent-almanac -a claude-code
插件命令备选方式
/plugin add https://github.com/pjt222/agent-almanac
Git 克隆备选方式
git clone https://github.com/pjt222/agent-almanac.git ~/.claude/skills/adaptic

在 Claude Code 中复制并粘贴此命令以安装该技能

技能文档

Adaptic

Compose the 5-step synoptic cycle to achieve panoramic synthesis across multiple domains. Where sequential analysis produces compromise ("a little of each"), the synoptic cycle produces integration — a unified understanding that holds all domains simultaneously and finds the emergent whole.

When to Use

  • A problem genuinely spans 3+ domains and the interactions between domains matter more than depth in any one
  • Sequential analysis (polymath style) has been tried but the synthesis feels like compromise rather than integration
  • Existing approaches feel like "a little of each" rather than a unified vision
  • Before major architectural decisions affecting multiple stakeholders
  • When domain experts disagree and the resolution lies between their perspectives, not within any one

When NOT to Use

  • Single-domain problems — use the domain agent directly
  • Well-understood trade-offs where polymath-style sequential analysis suffices
  • Self-care or wellness contexts — use the tending team instead
  • When speed matters more than depth — the full cycle requires sustained attention

Inputs

  • Required: The problem or question requiring multi-domain synthesis
  • Optional: Explicit list of domains to hold (default: auto-detect from problem context)
  • Optional: Depth setting — light, standard, or deep (default: standard)
  • Optional: Expression form — narrative, diagram, table, or recommendation (default: auto)

Configuration

settings:
  depth: standard          # light (skip meditate), standard, deep (extended perceive)
  domains: auto            # auto-detect or explicit list
  expression_form: auto    # narrative, diagram, table, recommendation

Procedure

Step 1: Clear — Empty the Workspace

Run the meditate skill to clear prior context, assumptions, and single-domain bias.

  1. Execute the full meditate procedure: prepare, anchor, observe distractions, close
  2. Pay special attention to domain bias — the tendency to frame the problem through whichever domain was most recently active
  3. Clear any premature solutions that arrived before the full picture was visible
  4. If depth: light is set, abbreviate to a brief context-clearing pause rather than the full meditation

Got: The workspace is empty. No domain has priority. No solution has been pre-selected. The agent is in a neutral, receptive state ready to hold multiple perspectives simultaneously.

If fail: If a particular domain keeps asserting itself as "the real problem," name that bias explicitly: "I notice I am framing this as primarily a [domain] problem." Naming the bias loosens its grip. If clearing fails entirely, the problem may genuinely be single-domain — reconsider whether the synoptic cycle is needed.

Step 2: Open — Enter Panoramic Mode

Run the expand-awareness skill to shift from narrow focus to wide-field perception.

  1. Inventory all domains relevant to the problem — do not pre-filter or rank them
  2. For each domain, note its core concerns, constraints, and values without evaluating
  3. Soften the focus: hold all domains in awareness simultaneously rather than cycling through them one at a time
  4. Resist the pull to "start solving" — this step is purely about opening the field of view
  5. If domains were provided explicitly in the inputs, use those as the starting set but remain open to discovering additional relevant domains

Got: A panoramic field is open. All relevant domains are held in awareness simultaneously. The agent can sense the full landscape without zooming into any single domain. The feeling is spacious rather than overwhelming.

If fail: If the domain list feels incomplete, ask: "What perspective is missing that would change the picture?" If simultaneous awareness collapses into sequential scanning (domain A, then B, then C), slow down — the goal is to hold the whole field, not to tour its parts. If more than 7 domains are active, group related domains into clusters to reduce cognitive load while maintaining breadth.

Step 3: Perceive — Notice Cross-Domain Patterns

While maintaining panoramic awareness, run observe and awareness to notice patterns, tensions, and resonances across all visible domains.

  1. Hold the panoramic field open from Step 2 — do not narrow focus
  2. Run observe to notice what is present: what patterns repeat across domains? what tensions exist between domains? what resonances connect seemingly unrelated concerns?
  3. Run awareness to notice what is not being seen: which domains are being subtly ignored? where are blind spots? what assumptions are operating below the surface?
  4. Record cross-domain observations without interpreting them yet:
    • Tensions: where domains pull in opposite directions
    • Resonances: where domains reinforce or echo each other
    • Gaps: where no domain addresses a concern that the whole picture reveals
    • Surprises: where a domain contributes something unexpected to the picture
  5. If depth: deep is set, extend this step — cycle through observe and awareness multiple times, allowing subtler patterns to surface

The critical discipline: perceive across all domains simultaneously, not each domain in turn. Sequential perception loses the cross-domain patterns that are the entire point of the synoptic cycle.

Got: A rich set of cross-domain observations — tensions, resonances, gaps, and surprises. These observations span the boundaries between domains rather than living within any single one. The agent has noticed something that would not be visible from any single domain's perspective.

If fail: If observations are all within single domains ("in domain A, I notice X"), the panoramic field has collapsed. Return to Step 2 and re-open. If no cross-domain patterns emerge, the problem may not require synoptic treatment — it may be genuinely decomposable into independent domain problems. If the perceive step produces an overwhelming number of observations, prioritize tensions (they are where integration happens).

Step 4: Integrate — Form the Emergent Whole

Run the integrate-gestalt skill to synthesize cross-domain observations into a unified understanding.

  1. Map the tensions identified in Step 3 — do not resolve them prematurely; hold them as creative constraints
  2. Find the figure: what unified understanding emerges when all observations are held together? This is not a compromise or average — it is a new pattern that includes but transcends the individual domain perspectives
  3. Test the whole: does the integrated understanding honor each domain's core concerns? Does it resolve tensions or merely paper over them?
  4. Name the insight in one clear statement — if it cannot be stated simply, the integration is not yet complete
  5. Verify that the insight is genuinely emergent: could it have been reached by analyzing domains sequentially? If yes, the synoptic cycle added no value and sequential analysis would have sufficed

Got: A single integrated understanding that holds all domains simultaneously. The insight feels like discovery rather than construction — it emerged from the whole rather than being assembled from parts. Each domain's core concerns are honored, and the tensions between domains are resolved rather than compromised.

If fail: If integration produces "a little of each domain" rather than a unified whole, the gestalt has not formed. Return to Step 3 and look for the tensions that are being avoided — integration happens through tension, not around it. If no gestalt forms after extended effort, decompose: find the 2-3 domains with the strongest tensions and integrate those first, then expand.

Step 5: Express — Communicate the Integrated Understanding

Run the express-insight skill to communicate the synthesis to the intended audience.

  1. Assess the audience: what domains are they familiar with? what framing will make the integrated insight accessible?
  2. Choose the expression form (or use the one specified in inputs):
    • Narrative: for audiences that need to understand the journey from parts to whole
    • Diagram: for audiences that need to see structural relationships
    • Table: for audiences that need to compare domain perspectives systematically
    • Recommendation: for audiences that need an actionable decision
  3. Express the integrated understanding with transparency: show which domains contributed, where tensions were resolved, and what the emergent insight adds beyond any single perspective
  4. Invite challenge: explicitly note which aspects of the integration are strongest and which are most speculative

Got: A clear, well-formed expression of the integrated understanding that is accessible to the intended audience. The expression shows its work — the audience can see how domain perspectives contributed to the whole. The form matches the audience's needs.

If fail: If the expression feels like a list of domain perspectives rather than an integrated whole, the insight from Step 4 has been lost in translation. Return to the one-statement summary from Step 4 and build the expression outward from that center. If the audience framing is wrong, ask: "Who needs this and what decision does it inform?"

Validation

  • Step 1 (Clear) was executed — prior context and domain bias were explicitly released
  • Step 2 (Open) produced a panoramic field holding 3+ domains simultaneously
  • Step 3 (Perceive) identified cross-domain patterns (not within-domain observations)
  • Step 4 (Integrate) produced a single emergent insight that transcends any individual domain
  • Step 5 (Express) communicated the insight in a form appropriate to the audience
  • The final output could not have been produced by sequential single-domain analysis
  • Each domain's core concerns are honored in the integrated understanding
  • Tensions between domains were resolved through integration, not compromise

Pitfalls

  • Sequential masquerading as simultaneous: Cycling through domains one at a time and then stapling the results together is not synoptic perception. The test: did the cross-domain interactions produce something new, or is the output a concatenation of domain analyses?
  • Premature integration: Jumping to a synthesis before the panoramic field has fully opened. Steps 2 and 3 build the perceptual foundation that makes genuine integration possible — rushing them produces shallow synthesis.
  • Compromise instead of emergence: Averaging domain perspectives ("50% security, 50% usability") is compromise, not integration. True integration finds a frame where both concerns are fully met, or it honestly names the irreducible trade-off.
  • Overuse on single-domain problems: Not every problem needs panoramic synthesis. If the problem lives cleanly in one domain, synoptic treatment adds overhead without value. The "When NOT to Use" criteria exist for a reason.
  • Losing the insight in expression: Step 4 produces a clear gestalt, but Step 5 fragments it back into a domain-by-domain list. Keep the integrated insight as the center of expression; domain details are supporting evidence, not the main structure.
  • Domain inflation: Artificially expanding the domain count to justify synoptic treatment. Three genuinely relevant domains produce better synthesis than seven domains where four are peripheral.

Related Skills

  • meditate — Step 1 of the cycle; clears context and establishes neutral starting state
  • expand-awareness — Step 2 of the cycle; shifts from narrow focus to panoramic perception
  • observe — used in Step 3; notices what is present across the field
  • awareness — used in Step 3; notices what is not being seen, reveals blind spots
  • integrate-gestalt — Step 4 of the cycle; forms the emergent whole from cross-domain patterns
  • express-insight — Step 5 of the cycle; communicates the integrated understanding

GitHub 仓库

pjt222/agent-almanac
路径: i18n/caveman-lite/skills/adaptic
0
agentsagentskillsai-assisted-developmentclaude-codeskillsteams

相关推荐技能

content-collections

Content Collections 是一个 TypeScript 优先的构建工具,可将本地 Markdown/MDX 文件转换为类型安全的数据集合。它专为构建博客、文档站和内容密集型 Vite+React 应用而设计,提供基于 Zod 的自动模式验证。该工具涵盖从 Vite 插件配置、MDX 编译到生产环境部署的完整工作流。

查看技能

polymarket

这个Claude Skill为开发者提供完整的Polymarket预测市场开发支持,涵盖API调用、交易执行和市场数据分析。关键特性包括实时WebSocket数据流,可监控实时交易、订单和市场动态。开发者可用它构建预测市场应用、实施交易策略并集成实时市场预测功能。

查看技能

creating-opencode-plugins

该Skill帮助开发者创建OpenCode插件,用于接入命令、文件、LSP等25+种事件。它提供了插件结构、事件API规范和JavaScript/TypeScript实现模式,适合需要拦截操作、扩展功能或自定义事件处理的场景。开发者可通过它快速构建响应式模块来增强OpenCode AI助手的能力。

查看技能

sglang

SGLang是一个专为LLM设计的高性能推理框架,特别适用于需要结构化输出的场景。它通过RadixAttention前缀缓存技术,在处理JSON、正则表达式、工具调用等具有重复前缀的复杂工作流时,能实现极速生成。如果你正在构建智能体或多轮对话系统,并追求远超vLLM的推理性能,SGLang是理想选择。

查看技能