[{"content":"Agent Skills #Repository-local Agent Skills that guide AIMS automation, OKF curation, site generation, and PR review.\nRepository facts #Repository-local Agent Skills describe repeatable workflows for market analysis, CFD instrument updates, local QA, PR feedback triage, and OKF maintenance. The OKF author, curator, site, and PR-review skills point agents at canonical okf/ edits and generated content/knowledge/ validation.\nAgent workflows must keep generated knowledge separate from canonical OKF and must not invent numeric market facts.\nSource-of-truth boundary #AIMS keeps numeric market facts, scores, ranks, dates, risk gates, and data availability in generated artifacts and validated reports. This OKF concept captures durable repository knowledge only.\nRelated concepts # Architecture Operational Recovery Citations # .agents/skills/local-qa/SKILL.md .agents/skills/market-analysis/SKILL.md .agents/skills/update-cfd-instruments/SKILL.md .agents/skills/aims-okf-author/SKILL.md ","date":"January 1, 0001","permalink":"https://dceoy.github.io/aims/knowledge/concepts/agent-skills/","section":"AIMS Knowledge Base","summary":"","title":"Agent Skills"},{"content":"Financial market analysis results are collected here as static pages.\nAdd new analysis reports under content/results/; Hugo will render them into the generated site/ directory.\n","date":null,"permalink":"https://dceoy.github.io/aims/","section":"AIMS","summary":"","title":"AIMS"},{"content":"AIMS Architecture #Repository-level architecture for the AIMS analysis pipeline, Hugo site, OKF source, and generated knowledge pages.\nRepository facts #AIMS combines Python automation with a Hugo static site. Python scripts live under .agents/skills/, tests live in tests/, schemas live under data/schema/, and Hugo source content lives under content/. Generated static output is written to the ignored site/ directory.\nThe OKF layer adds okf/ as the durable knowledge source while preserving existing source-of-truth boundaries: daily analysis JSON remains under data/analysis/, public reports remain under content/results/, and content/knowledge/ is regenerated from OKF.\nSource-of-truth boundary #AIMS keeps numeric market facts, scores, ranks, dates, risk gates, and data availability in generated artifacts and validated reports. This OKF concept captures durable repository knowledge only.\nRelated concepts # Data Sources Report Generation Publication Workflow Citations # AGENTS.md README.md ","date":"January 1, 0001","permalink":"https://dceoy.github.io/aims/knowledge/concepts/architecture/","section":"AIMS Knowledge Base","summary":"","title":"AIMS Architecture"},{"content":"AIMS Knowledge Base #The okf/ tree is the canonical repository-native source for durable AIMS architecture, operations, and methodology knowledge. Hugo pages under content/knowledge/ are generated shadow content and must not be hand-edited.\nConcepts # Agent Skills AIMS Architecture Data Sources Instrument Master Operational Recovery Publication Workflow Report Generation Scoring Methodology Log # Knowledge log ","date":null,"permalink":"https://dceoy.github.io/aims/knowledge/","section":"AIMS Knowledge Base","summary":"","title":"AIMS Knowledge Base"},{"content":"AIMS OKF Log #2026-06-16 # Established okf/ as the canonical OKF v0.1 knowledge source for durable AIMS concepts. Added generated content/knowledge/ shadow content as Hugo publication output, not a canonical authoring location. Seeded initial concepts for architecture, data sources, instrument master data, scoring methodology, report generation, publication, recovery, and Agent Skills. ","date":"January 1, 0001","permalink":"https://dceoy.github.io/aims/knowledge/logs/log/","section":"AIMS Knowledge Base","summary":"","title":"AIMS OKF Log"},{"content":"Published market analysis reports appear in this section.\n","date":null,"permalink":"https://dceoy.github.io/aims/results/","section":"Analysis Results","summary":"","title":"Analysis Results"},{"content":"Data Sources #Authoritative market and instrument data sources used by AIMS and the boundaries for durable knowledge updates.\nRepository facts #AIMS fetches daily OHLCV history from Stooq using its free CSV download API. Configured Stooq symbols are maintained in data/stooq_symbols.txt.\nThe CFD instrument master is a separate data source maintained in data/cfd_instruments.csv. It is refreshed by the weekly updater workflow and validated before the daily market analysis workflow runs.\nSource-of-truth boundary #AIMS keeps numeric market facts, scores, ranks, dates, risk gates, and data availability in generated artifacts and validated reports. This OKF concept captures durable repository knowledge only.\nRelated concepts # Instrument Master Scoring Methodology Citations # OPERATIONS.md .github/workflows/daily-market-analysis.yml .github/workflows/update-cfd-instruments.yml ","date":"January 1, 0001","permalink":"https://dceoy.github.io/aims/knowledge/concepts/data-sources/","section":"AIMS Knowledge Base","summary":"","title":"Data Sources"},{"content":"Instrument Master #Canonical CFD instrument reference data and validation responsibilities in AIMS.\nRepository facts #The CFD instrument master is stored in data/cfd_instruments.csv and sourced from GMO Click Securities and Rakuten Securities CFD lineup pages. Broker ticker symbols are maintained separately from Stooq symbols, with mappings in data/mappings/cfd_ticker_mappings.csv.\nThe master CSV is validated against data/schema/cfd_instruments.schema.json after updates and before daily analysis.\nSource-of-truth boundary #AIMS keeps numeric market facts, scores, ranks, dates, risk gates, and data availability in generated artifacts and validated reports. This OKF concept captures durable repository knowledge only.\nRelated concepts # Data Sources Scoring Methodology Citations # OPERATIONS.md data/schema/cfd_instruments.schema.json .agents/skills/update-cfd-instruments/SKILL.md ","date":"January 1, 0001","permalink":"https://dceoy.github.io/aims/knowledge/concepts/instrument-master/","section":"AIMS Knowledge Base","summary":"","title":"Instrument Master"},{"content":"Operational Recovery #Runbook-oriented knowledge for restoring AIMS analysis, validation, and publication workflows.\nRepository facts #Operational recovery starts from the failing workflow step: Stooq fetch warnings are non-fatal per symbol, artifact validation failures should be fixed in the generated JSON or generator inputs, and site build failures should be reproduced locally with Hugo.\nManual recovery uses the documented scripts and validations rather than editing generated outputs by hand.\nSource-of-truth boundary #AIMS keeps numeric market facts, scores, ranks, dates, risk gates, and data availability in generated artifacts and validated reports. This OKF concept captures durable repository knowledge only.\nRelated concepts # Publication Workflow Agent Skills Citations # OPERATIONS.md .agents/skills/local-qa/SKILL.md ","date":"January 1, 0001","permalink":"https://dceoy.github.io/aims/knowledge/concepts/operational-recovery/","section":"AIMS Knowledge Base","summary":"","title":"Operational Recovery"},{"content":"Publication Workflow #How CI/CD, pull requests, and GitHub Pages publish AIMS reports and generated OKF shadow content.\nRepository facts #The daily analysis workflow validates the CFD master, generates JSON artifacts, validates the artifact, generates a Hugo report, builds the Hugo site for validation, and opens a pull request for review.\nThe CI workflow deploys the Hugo site to GitHub Pages after linting, type checking, and tests pass. The OKF validation job checks generated knowledge content drift before deployment.\nSource-of-truth boundary #AIMS keeps numeric market facts, scores, ranks, dates, risk gates, and data availability in generated artifacts and validated reports. This OKF concept captures durable repository knowledge only.\nRelated concepts # Architecture Operational Recovery Citations # OPERATIONS.md .github/workflows/daily-market-analysis.yml .github/workflows/ci.yml ","date":"January 1, 0001","permalink":"https://dceoy.github.io/aims/knowledge/concepts/publication-workflow/","section":"AIMS Knowledge Base","summary":"","title":"Publication Workflow"},{"content":"Report Generation #How deterministic JSON analysis artifacts become public Hugo market analysis reports.\nRepository facts #The report generator reads data/analysis/YYYY-MM-DD.json and writes Hugo Markdown reports to content/results/YYYY-MM-DD-market-analysis.md. The generator is deterministic: identical JSON input produces identical Markdown output and the report timestamp comes from the artifact.\nOKF-generated content/knowledge/ pages are separate from content/results/ reports and do not replace daily report generation.\nSource-of-truth boundary #AIMS keeps numeric market facts, scores, ranks, dates, risk gates, and data availability in generated artifacts and validated reports. This OKF concept captures durable repository knowledge only.\nRelated concepts # Architecture Publication Workflow Citations # OPERATIONS.md README.md tests/golden/2024-01-01-market-analysis.md ","date":"January 1, 0001","permalink":"https://dceoy.github.io/aims/knowledge/concepts/report-generation/","section":"AIMS Knowledge Base","summary":"","title":"Report Generation"},{"content":"Scoring Methodology #How AIMS treats analysis artifacts, score semantics, ranking outputs, and risk gates as generated numeric facts.\nRepository facts #AIMS computes cross-sectional rankings from daily OHLCV-derived features and produces a composite score from percentile ranks. Risk gates mark unreliable instruments, which remain in output but are ranked below reliable instruments.\nOKF concepts may summarize durable methodology, but numeric market facts, scores, ranks, dates, risk gates, and availability remain authoritative only in generated analysis artifacts and validated report outputs.\nSource-of-truth boundary #AIMS keeps numeric market facts, scores, ranks, dates, risk gates, and data availability in generated artifacts and validated reports. This OKF concept captures durable repository knowledge only.\nRelated concepts # Data Sources Report Generation Citations # OPERATIONS.md data/schema/analysis.schema.json AGENTS.md ","date":"January 1, 0001","permalink":"https://dceoy.github.io/aims/knowledge/concepts/scoring-methodology/","section":"AIMS Knowledge Base","summary":"","title":"Scoring Methodology"}]