AgentSkillsCN

extract

在不执行完整三阶段分析的情况下，从源数据中提取声明。适用于快速抓取声明或对相关来源进行辅助分析。

SKILL.md
--- frontmatter
name: extract
description: "Extract claims from a source without performing full 3-stage analysis. Use for quick claim harvesting or supporting sources."
<!-- GENERATED FILE - DO NOT EDIT DIRECTLY --> <!-- Source: integrations/_templates/ + _config/skills.yaml --> <!-- Regenerate: make assemble-skills -->

Quick Claim Extraction (Codex)

Extract claims from a source without performing full 3-stage analysis. Use for quick claim harvesting or supporting sources.

Invocation

code
$extract <source>

Note: Codex reserves /... for built-in commands. Use $extract instead.

Extract claims from a source without performing full 3-stage analysis. Use for quick claim harvesting or when sources are supporting evidence rather than primary subjects.

When to Use

  • •Quick extraction of claims from supporting sources
  • •Building cross-reference evidence for existing claims
  • •Initial triage before deciding on full analysis
  • •Sources that are data-heavy rather than argument-heavy

Output Format

yaml
claims:
  - id: "DOMAIN-YYYY-NNN"
    text: "[Precise claim statement]"
    type: "[F/T/H/P/A/C/S/X]"
    domain: "[DOMAIN]"
    evidence_level: "E[1-6]"
    credence: 0.XX
    source_ids: ["[source-id]"]

Claim Types

TypeSymbolDefinition
Fact[F]Empirically verified, consensus reality
Theory[T]Coherent explanatory framework with empirical support
Hypothesis[H]Testable proposition, awaiting evidence
Prediction[P]Future-oriented claim with specified conditions
Assumption[A]Underlying premise (stated or unstated)
Counterfactual[C]Alternative scenario for comparison
Speculation[S]Unfalsifiable or untestable claim
Contradiction[X]Identified logical inconsistency

Evidence Hierarchy

Use this hierarchy to rate strength of evidential support for claims.

LevelStrengthDescriptionCredence Range
E1Strong EmpiricalSystematic review, meta-analysis, replicated experiments0.9-1.0
E2Moderate EmpiricalSingle peer-reviewed study, official statistics0.6-0.8
E3Strong TheoreticalExpert consensus, working papers, preprints0.5-0.7
E4Weak TheoreticalIndustry reports, credible journalism0.3-0.5
E5Opinion/ForecastPersonal observation, anecdote, expert opinion0.2-0.4
E6UnsupportedPure speculation, unfalsifiable claims0.0-0.2

Domain Codes

CodeDescription
TECHTechnology, AI capabilities, tech trajectories
LABOREmployment, automation, human work
ECONValue theory, pricing, distribution, ownership
GOVGovernance, policy, regulation
SOCSocial structures, culture, behavior
RESOURCEScarcity, abundance, allocation
TRANSTransition dynamics, pathways
GEOInternational relations, state competition
INSTInstitutions, organizations
RISKRisk assessment, failure modes
METAClaims about the framework/analysis itself

Notes

  • •Quick extractions should be labeled with Analysis Depth: quick in the analysis file
  • •Minimum requirements: Metadata, Legends, Claim Summary table, Claims YAML
  • •Full Stage 2 evaluation tables (verification, counterevidence, tensions, persuasion) are optional for quick extractions
  • •If a source proves more significant, upgrade to full analysis with $check --continue

Related Skills

  • •$check
  • •$analyze