AGENT SKILLS · ミッション装備

実際に使われている Agent Skills

公開リポジトリの検証済み Skills を比較し、機能、内容、導入状況、出典を確認してから追加できます。

掲載範囲: 検証済み
Skills: 22,797
リポジトリ: 3512
最終同期: 2026/11/01

22,797 Skills

extract-empirical-regularity

Extract a repeatable empirical pattern from observations, state its support and exceptions, and generalize cautiously without importing an unsupported mechanism.

1時間の導入数: 1
GitHub Stars: 505
更新日: 9月29日

Generate testable hypotheses from theory, empirical regularity, anomaly, or explicit explanatory competition. Deductive, inductive, abductive, and competing-hypothesis generation are modes; discriminating predictions and comparison are activated in competing mode.

1時間の導入数: 1
GitHub Stars: 505
更新日: 9月29日

formulate-research-question

Convert a hypothesis into a precise research question with fit-for-purpose framework, scope, feasibility, and success criteria.

1時間の導入数: 1
GitHub Stars: 505
更新日: 9月29日

generate-competing-hypotheses

Generate mechanism-distinct explanations rather than superficial paraphrases.

1時間の導入数: 1
GitHub Stars: 505
更新日: 9月29日

generate-counterexample

Generate a concrete counterexample candidate that satisfies the target claim's stated scope while violating its conclusion or claimed structural property; record whether exclusion would be principled or ad hoc.

1時間の導入数: 1
GitHub Stars: 505
更新日: 9月29日

list-undesirable-effects

List observable undesirable effects with evidence and severity as starting points for constraint/root-cause analysis.

1時間の導入数: 1
GitHub Stars: 505
更新日: 9月29日

map-stakeholder-jobs

Represent stakeholder jobs, desired outcomes, pains, gains, dependencies, and conflicts relevant to the research target.

1時間の導入数: 1
GitHub Stars: 505
更新日: 9月29日

map-threat-surface

Enumerate attackable dimensions of an artifact and rate exposure/accessibility.

1時間の導入数: 1
GitHub Stars: 505
更新日: 9月29日

negate-claim

Produce a scope-preserving logical negation of a claim or assumption.

1時間の導入数: 1
GitHub Stars: 505
更新日: 9月29日

normalize-gap

Normalize heterogeneous gap records into a comparable schema before decision operations.

1時間の導入数: 1
GitHub Stars: 505
更新日: 9月29日

parse-patent-claim

Parse independent/dependent claim structure, extract elements, and map claim features to technical functions/domains.

1時間の導入数: 1
GitHub Stars: 505
更新日: 9月29日

plan-effect-size

Choose effect-size estimands and required conversions/calculation rules appropriate to the outcomes and design.

1時間の導入数: 1
GitHub Stars: 505
更新日: 9月29日

problem-reframing

Change the problem representation when the current frame is limiting: dominant-frame escape, stakeholder/worldview rotation, forced perspective shifts, dialectical/polarity reframing, or abstraction/scope shifts.

1時間の導入数: 1
GitHub Stars: 505
更新日: 9月29日

represent-mechanism-edge

Represent a directed mechanism edge with pathway, sign/direction, assumptions, falsifier, strength, and evidence provenance.

1時間の導入数: 1
GitHub Stars: 505
更新日: 9月29日

resolve-inventive-contradiction

Turn a design or research contradiction into targeted transformations using inventive-principle and separation reasoning, then evaluate whether the transformed configuration resolves the contradiction without creating a worse one.

1時間の導入数: 1
GitHub Stars: 505
更新日: 9月29日

select-critical-case

Select most-likely, least-likely, or otherwise inference-maximizing cases that can decisively support/refute a claim under explicit case-selection logic.

1時間の導入数: 1
GitHub Stars: 505
更新日: 9月29日

select-experimental-baseline

Select fair SOTA/simple/oracle/control baselines with provenance and rationale matched to the experimental claim.

1時間の導入数: 1
GitHub Stars: 505
更新日: 9月29日

select-from-frontier

Select a preferred frontier point/portfolio and state the sacrificed alternatives/tradeoffs.

1時間の導入数: 1
GitHub Stars: 505
更新日: 9月29日

select-inventive-principle

Select one or more inventive principles appropriate to a contradiction or transformation objective and state why each principle applies.

1時間の導入数: 1
GitHub Stars: 505
更新日: 9月29日

select-next-pair

Select the next comparison pair to maximize expected information gain.

1時間の導入数: 1
GitHub Stars: 505
更新日: 9月29日

select-statistical-method

Select statistical inference/estimation method from design, distribution, sample size, pairing, multiplicity, and decision objective.

1時間の導入数: 1
GitHub Stars: 505
更新日: 9月29日

sensitivity-analysis

Quantify which parameters, assumptions, interactions, or uncertainties dominate the conclusion; select Morris/Sobol/perturbation/Monte-Carlo as modes.

1時間の導入数: 1
GitHub Stars: 505
更新日: 9月29日