Каталог навыков агентов

AI thinking partner for strategy and planning. Reads the current work state, analyzes progress, and helps you plan the n...

Script a teacher think-aloud demonstrating expert reasoning processes for a specific task. Use when modelling problem-so...

Forces explicit reasoning before writing any code. Surfaces assumptions, manages confusion, and prevents hallucination b...

Zoom out from building/executing mode and think big about how something you have already built (or are deep in) could gr...

Strategic thinking framework — analyze whether something is worth doing, what the risks are, and what the long-term impa...

Good-faith divergent brainstorming for achieving a goal. Validates assumptions, then spawns brainstormers running differ...

当用户需要多角度分析问题、深入讨论话题、验证想法、发现思维盲区、评估决策风险、审视方案时，使用此技能。自动组建多视角讨论团（通常5人），通过独立思考、基于内容的交叉讨论和结构化总结，帮助用户形成更全面的判断。触发场景包括但不限于：用户说"分...

Adversarial deliberation process for making decisions. Spawns advocate agents for each option who argue their cases befo...

Abductive reasoning about causes. Takes a phenomenon, separates observations from interpretations, then spawns diagnosti...

Get unstuck on a problem by deliberately re-framing it through a different lens — a child's, an outsider's, another indu...

4-phase structured analysis wrapper for vibe pre-routing (compatible with claude-code-settings think-harder semantics).

Use when analyzing multiple files, auditing a codebase, scanning dependencies, or counting lines -- one Bash pipeline be...

Apply the GPT-5.6 Sol execution-intelligence framework to any complex task. Use when the user asks the agent to "think l...

Retrospective learning from a completed experience. Takes a project, incident, decision that played out, or time period;...

Problem redefinition before problem solving. Takes a stated problem, extracts its premises, then spawns reframers applyi...

Good-faith adversarial scrutiny of an idea or plan. Spawns critical skeptics across relevant lenses, pairs them with an...

Collaborative exploration and reasoning frameworks. Use when ideas need shaping, decisions need analysis, or requirement...

Forces Claude to pause before picking an implementation approach and ask: "Is there a cleverer, cheaper way?" Triggers w...

When the user wants to review past decisions, understand why something was built a certain way, audit decision quality,...

Use this skill whenever the user brings a raw idea, concept, or "I want to build X" — OR when invoked on an existing pro...

Transform brain dumps, rough ideas, and vague requirements into professional, well-structured prompts following Claude's...

Cognition orchestrator — analyze canvases, distill fears via /distill subagent, run gap analysis, optional cross-user sy...

The working method for every nontrivial task: frame the objective, ground load-bearing facts, converge on a route, execu...

Decision-framing workflow for turning an ambiguous problem or open choice into a recommendation, experiment, or concrete...

Recognize Senge's Systems Archetypes to diagnose recurring organizational and technical problems, identify why fixes kee...

Update beliefs systematically based on new evidence using probabilistic reasoning. Use when estimating probabilities, le...

Use when search or investigation could run forever. Set an explicit good-enough threshold first, then stop at the first...

Use when a specific claim may lack grounding. Check evidence boundary, size wrongness cost, then answer, fetch, or absta...

Deep engineering analysis before implementation (Phase 2 of 9-Phase Feature Cycle). Challenges assumptions, identifies u...

When the right response mode is unclear, classify the cause-effect domain first; decompose disorder.

Systematic checklist to identify and counteract cognitive biases in decision-making. Use before major decisions, when ev...

Engages structured analysis to explore multiple perspectives and context dependencies before responding. Use when users...

Apply Kahneman's Dual-Process Theory to recognize when to trust intuition vs engage deliberate analysis. Use for high-st...

Under genuine uncertainty with no reliable forecast, inventory means, cap downside at affordable loss, act for commitmen...

Analyze systems using Donella Meadows' feedback loop framework to identify reinforcing loops, balancing loops, delays, a...

Make order-of-magnitude estimates for unknown quantities by decomposing into known or estimable factors. Use for capacit...

When a constraint is treated as fixed, separate physics from convention, keep only independently supported primitives, a...

When a fault is localized and the proximate cause is known but the systemic root is not, chain evidence-linked whys with...

Use this when complex problem-solving, root cause analysis, strategic decision-making, or systematic thinking is needed....

PROACTIVELY invoke this skill for any task, problem, or decision requiring structured thought, fresh perspectives, or 'o...

Collection of mental models and decision-making frameworks (first principles, 5 whys, SWOT, Occam's razor, Pareto, inver...

Approach problems backward by identifying paths to failure, then systematically avoiding them. Use for risk identificati...

Deciding what to build or why adoption fails. Recover the progress users hire a solution for under a circumstance, then...

Use when a selective defect needs IS/IS-NOT difference analysis or a consequential option choice needs must/want weighti...

Identify where small changes can have large effects using Donella Meadows' hierarchy of system intervention points. Use...

Use when longevity of a non-perishable option matters. Treat survival duration as a remaining-life prior, then check dom...

When a claim, doc, test, metric, or assumption conflicts with observed behavior, stop theorizing from the map and verify...

When provisioning, setting a limit, or committing an estimate under uncertainty, size a buffer to residual error and the...

Use when the user describes a problem, decision or stuck situation but names no method: "how should I approach this?", "...

When one mental model leaves a material blind spot on a multi-domain or high-stakes problem, sequence complementary mode...

When unsure which thinking skill fits, map domain and problem type, then return NONE or one primary skill by default (at...

Choose the right mental model for the problem at hand. Use when facing new problems, when current approaches fail, or wh...

Extended thinking modes for deeper analysis. Use when user says 'think', 'think hard', 'ultrathink', 'super think', or '...

Apply parsimony principle to prefer simpler explanations with fewer assumptions. Use for hypothesis selection in debuggi...

Use under time pressure when the situation is still changing and you must act before certainty — cycle Observe→Orient→De...

Before committing scarce time, people, or money, name the best forgone use of those resources and the value delta of the...

A contract for what the agent does when a long, messy, stream-of-consciousness ramble arrives (usually voice dictation):...

Collaborative thinking partner for exploring complex problems through questioning

A deterministic thinking partner that challenges assumptions and applies mental models to sharpen decisions, solve probl...

Before committing to a plan or launch, assume it already failed and reason backward through concrete causes — convert fa...

Use when forecasting, estimating, or sizing risk — anchor on base rates, give ranges, update prior→likelihood→posterior...

For authorized security review of code, auth, or APIs you control, model the attacker, map the attack surface, and repor...

Project to your future self and ask what you would regret not doing. Use for career decisions, strategic pivots, risk-ta...

Before heavy deliberation, classify the decision as cheap or costly to undo; decide two-way doors fast and stage one-way...

When a symptom has several plausible causes, rank falsifiable hypotheses and run the cheapest discriminating observation...

When a change has effects past the immediate fix—incentives, scale, feedback—trace consequence chains with timing and pr...

When a request is vague, assumption-laden, or "obvious," ask the few load-bearing questions that expose hidden requireme...

Before rejecting a proposal or reflexively agreeing, build the strongest faithful opposing case, state agreement conditi...

When behavior is emergent across components—fixes elsewhere break, loops/delays dominate—map boundary, stocks/flows, fee...

When throughput or latency is pipeline-limited, identify the single binding constraint and exploit, subordinate, elevate...

When a real test is too rare, large, or irreversible, run a controlled counterfactual: isolate one variable, fix conditi...

Structured thinking patterns for agent self-reflection. Includes think-about-collected-information (validate research),...

When two design requirements seem mutually exclusive, name the contradiction, separate conflicting states, then invent a...

Use when the reflex is to add a feature, layer, or process. Prefer removing harmful or nonessential elements first, with...

無駄なスキル増殖を防ぎつつ、必要な時に確実に使えるスキルだけを増やす ことです。

After Claude self-pitfall + Codex on a ship-worthy/architecture/RT/security/contract change: run a third external model...

Produces the scope document for an external diligence provider -- the questions to answer, the deliverable format, the m...

Best practices for integrating third-party APIs and services in career_ios_backend. Automatically invoked when working w...

Assess the security and operational risk of third-party model APIs and vendors — covering due diligence requirements, AP...

Audits third-party Git repositories against upstream remotes for pending commits, tags, and patch status.

Third-party trust boundary analyzer: inventories every outbound HTTP/RPC call (fetch, axios, got, requests, curl, Invoke...

Thirdweb v5 SDK usage in AutoClaw. Use when working with wallet connection, social login, SIWE authentication, or thirdw...

Tactical 30-day business launch guide with copy-paste templates, hour-by-hour schedules, and revenue forecasts. For user...

A skill with a name that exceeds 64 characters.

Apply Seth Godin's marketing framework. Trigger on: "find my smallest viable audience", "position this offer", "earn tru...

Provides guidance on creating custom error types with thiserror, including proper derive macros, error messages, and sou...

Write systems code in the style of Ken Thompson, co-creator of Unix, C, Go, UTF-8, and Plan 9. Emphasizes radical simpli...

Analyze THOR's SQLite database (thor10.db/thor11.db) for performance tuning, scan timing, resume state, and delta compar...

THOR Lens workflows for forensic timeline analysis. A web UI that imports THOR v11 audit trail JSONL logs for interactiv...

Interpret THOR scan results and explain what findings mean. Use when the user pastes THOR log lines, shares a log file,...

- Be precise about thor-util verbs:

Write, package, and use THOR plugins to extend scanner functionality. THOR v11+ only.

Run THOR scans and propose the exact command line for Windows, Linux, or macOS. Use when the user wants to scan a host,...

This is the root skill. It routes requests to the right sub-skill and enforces a few global rules.

Troubleshoot THOR runs that are stuck, slow, failing to start, stopping early, or produce missing output. Use when the u...