[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"extension-skill-lllllllama-minimal-run-and-audit-de":3,"guides-for-lllllllama-minimal-run-and-audit":504,"similar-k175km7t1cvxck0ecpgxry0tbd86m157-de":505},{"_creationTime":4,"_id":5,"children":6,"community":7,"display":9,"evaluation":15,"identity":243,"isFallback":228,"parentExtension":248,"providers":249,"relations":255,"repo":258,"tags":500,"workflow":501},1778692706692.615,"k175km7t1cvxck0ecpgxry0tbd86m157",[],{"reviewCount":8},0,{"description":10,"installMethods":11,"name":13,"sourceUrl":14},"Vertrauenswürdige Ausführungs- und Reporting-Skill für die README-basierte Reproduktion von KI-Repositorien. Verwenden Sie diesen Skill, wenn die Aufgabe speziell darin besteht, Nachweise aus dem ausgewählten Smoke-Test oder dem dokumentierten Inferenz- oder Auswertungsbefehl zu erfassen oder zu normalisieren und standardisierte `repro_outputs/`-Dateien zu schreiben, einschließlich Patch-Notizen, wenn sich Repository-Dateien geändert haben. Nicht für Trainingsausführung, erstmalige Repositoireaufnahme, generische Umgebungs einrichtung, Paper-Suche, Zielauswahl oder reine End-to-End-Orchestrierung verwenden.",{"claudeCode":12},"lllllllama/ai-paper-reproduction-skill","minimal-run-and-audit","https://github.com/lllllllama/ai-paper-reproduction-skill",{"_creationTime":16,"_id":17,"extensionId":5,"locale":18,"result":19,"trustSignals":226,"workflow":241},1778692706692.6152,"kn76qte64a62t3mtceedc5dx9h86md9x","de",{"checks":20,"evaluatedAt":194,"extensionSummary":195,"features":196,"nonGoals":202,"promptVersionExtension":208,"promptVersionScoring":209,"purpose":210,"rationale":211,"score":212,"summary":213,"tags":214,"tier":220,"useCases":221},[21,26,29,32,36,39,44,48,51,54,58,62,65,69,72,75,78,81,84,87,91,95,99,103,107,110,113,116,120,123,126,129,132,135,138,142,146,150,153,157,160,163,166,169,173,176,179,182,185,188,191],{"category":22,"check":23,"severity":24,"summary":25},"Praktischer Nutzen","Problemrelevanz","pass","Die Beschreibung erklärt klar das Problem der Reproduktion von KI-Forschungsartefakten aus READMEs und benennt den Zielbenutzer (Teams, die KI-Workflows pflegen).",{"category":22,"check":27,"severity":24,"summary":28},"Alleinstellungsmerkmal","Der Skill bietet einen spezialisierten Workflow für die Reproduktion von KI-Repos und die Normalisierung von Nachweisen und geht über generische LLM-Funktionen hinaus, indem er sich auf spezifische Ausgabeformate und vertrauenswürdige Ausführungspfade konzentriert.",{"category":22,"check":30,"severity":24,"summary":31},"Produktionsbereitschaft","Das Skill-Bundle scheint für seinen angegebenen Zweck der Erfassung und Normalisierung von Ausführungsnachweisen vollständig zu sein, mit dokumentierten Skripten und einer klaren Berichterstattungsrichtlinie.",{"category":33,"check":34,"severity":24,"summary":35},"Umfang","Single Responsibility Principle","Der Skill `minimal-run-and-audit` konzentriert sich speziell auf die Erfassung und Normalisierung von Nachweisen aus dokumentierten Befehlen und passt in den breiteren KI-Workflow, ohne sich mit unrelateden Domänen zu überschneiden.",{"category":33,"check":37,"severity":24,"summary":38},"Qualität der Beschreibung","Die angezeigte Beschreibung spiegelt genau den Zweck des Skills wider: vertrauenswürdige Ausführung und Berichterstattung für die Reproduktion von KI-Repos, einschließlich klarer Anwendungsfälle und Nicht-Ziele.",{"category":40,"check":41,"severity":42,"summary":43},"Aufruf","Umfangreiche Tools","not_applicable","Dieser Skill scheint ein einzelner Single-Purpose-Einstiegspunkt zu sein und stellt keine mehreren Tools bereit. Die Kernlogik liegt wahrscheinlich im Skript `run_command.py`.",{"category":45,"check":46,"severity":24,"summary":47},"Dokumentation","Konfigurations- & Parameterreferenz","Die Betriebsparameter des Skills (Repo, Befehl, Timeout) sind über seinen Argument-Parser in `run_command.py` klar definiert und im SKILL.md und README dokumentiert.",{"category":33,"check":49,"severity":24,"summary":50},"Tool-Namensgebung","Der Haupteinstiegspunkt ist gut benannt (`minimal-run-and-audit`) und sein zugehöriges Skript (`run_command.py`) beschreibt seine Funktion.",{"category":33,"check":52,"severity":24,"summary":53},"Minimale I/O-Oberfläche","Das Skript `run_command.py` erfasst spezifische Ausführungsdetails (Rückgabecode, stdout, stderr, Dateisystemänderungen), die für seinen Zweck relevant sind, ohne übermäßige diagnostische Ausgaben.",{"category":55,"check":56,"severity":24,"summary":57},"Lizenz","Lizenznutzbarkeit","Das Repository ist unter der MIT-Lizenz lizenziert, einer permissiven Open-Source-Lizenz, die in der LICENSE-Datei klar angegeben ist.",{"category":59,"check":60,"severity":24,"summary":61},"Wartung","Aktualität der Commits","Der letzte Commit war am 9. Mai 2026, was innerhalb der letzten 90 Tage liegt und eine aktive Wartung anzeigt.",{"category":59,"check":63,"severity":42,"summary":64},"Abhängigkeitsverwaltung","Die bereitgestellten Skripte scheinen Standard-Python-Bibliotheken und git zu verwenden, ohne explizite Drittanbieter-Abhängigkeiten, die eine komplexe Verwaltung erfordern würden.",{"category":66,"check":67,"severity":24,"summary":68},"Sicherheit","Geheimnisverwaltung","Das Skript `run_command.py` führt einen bereitgestellten Befehl im Kontext eines Repositoriums aus und scheint keine Geheimnisse zu verarbeiten oder preiszugeben.",{"category":66,"check":70,"severity":24,"summary":71},"Injection","Das Skript `run_command.py` verwendet `shlex.split` zur Befehlsparsung und `subprocess` zur Ausführung, was direkte Injection-Risiken mindert. Git-Befehle werden ebenfalls sicher behandelt.",{"category":66,"check":73,"severity":24,"summary":74},"Transitive Lieferketten-Granaten","Das Skript führt lokale Befehle und Git-Operationen im Repository-Kontext aus und ruft zur Laufzeit keinen externen Code oder Daten ab.",{"category":66,"check":76,"severity":24,"summary":77},"Sandbox-Isolation","Das Skript operiert innerhalb des bereitgestellten Repository-Pfads, und seine Hauptaktionen sind die Ausführung von Befehlen und die Erfassung des Git-Status, die innerhalb der erwarteten Grenzen bleiben.",{"category":66,"check":79,"severity":24,"summary":80},"Sandbox-Escape-Primitive","Das Python-Skript verwendet standardmäßiges `subprocess.run` und scheint keine getrennten Prozesse oder Wiederholungsversuche für abgelehnte Aufrufe zu verwenden.",{"category":66,"check":82,"severity":24,"summary":83},"Datenexfiltration","Das Skript konzentriert sich auf die Erfassung lokaler Ausführungsdaten und des Git-Status; es scheint keine vertraulichen Daten zu lesen oder an Dritte zu übermitteln.",{"category":66,"check":85,"severity":24,"summary":86},"Versteckte Texttricks","Die gebündelten Skripte und SKILL.md-Dateien scheinen frei von versteckten Texttricks oder Verschleierungen zu sein, die das Modell irreführen sollen.",{"category":88,"check":89,"severity":24,"summary":90},"Hooks","Undurchsichtige Codeausführung","Das Python-Skript ist sauberer, lesbarer Quellcode und beinhaltet keine Verschleierung, Base64-Dekodierung oder Laufzeit-Skriptabrufe.",{"category":92,"check":93,"severity":24,"summary":94},"Portabilität","Strukturelle Annahme","Das Skript operiert innerhalb eines angegebenen Repository-Pfads und behandelt den Git-Status relativ zu diesem Pfad, ohne allgemeine Annahmen über die Projektstruktur des Benutzers zu treffen.",{"category":96,"check":97,"severity":24,"summary":98},"Vertrauen","Aufmerksamkeit für Issues","Mit 0 geöffneten und 0 geschlossenen Issues in den letzten 90 Tagen ist die Interaktion des Maintainers kein Indikator für Vernachlässigung.",{"category":100,"check":101,"severity":24,"summary":102},"Versionierung","Release-Management","Obwohl es kein Semver im SKILL.md Frontmatter oder in den GitHub Releases gibt, beziehen sich die Installationsanweisungen implizit über die Repo-URL auf einen bestimmten Commit/Tag, und es gibt eine CHANGELOG.md.",{"category":104,"check":105,"severity":24,"summary":106},"Codeausführung","Validierung","Das Skript `run_command.py` verwendet `shlex.split` zur Befehlsparsung und `subprocess.run` zur Ausführung mit expliziten Timeouts, was ein grundlegendes Maß an Eingabevalidierung bietet.",{"category":66,"check":108,"severity":24,"summary":109},"Ungeschützte destruktive Operationen","Die Hauptfunktion des Skripts besteht darin, einen gegebenen Befehl auszuführen und seine Ausgabe/Status zu berichten; es enthält keine inhärent destruktiven Operationen wie `rm -rf`.",{"category":104,"check":111,"severity":24,"summary":112},"Fehlerbehandlung","Das Skript `run_command.py` behandelt explizit `FileNotFoundError`, `subprocess.TimeoutExpired` und Nicht-Null-Rückgabecodes und liefert strukturierte Ausgaben und Protokolle.",{"category":104,"check":114,"severity":24,"summary":115},"Protokollierung","Das Skript erfasst stdout, stderr und git status, was als Audit-Log für das Ergebnis des ausgeführten Befehls und die Änderungen am Dateisystem dient.",{"category":117,"check":118,"severity":42,"summary":119},"Compliance","DSGVO","Der Skill arbeitet nicht mit personenbezogenen Daten; er führt Befehle in einem Code-Repository aus und erfasst Ausgaben und Dateistatus.",{"category":117,"check":121,"severity":24,"summary":122},"Zielmarkt","Der Skill arbeitet mit Code-Repositories und Standardbefehlsausführung, ohne regionale oder gerichtliche Einschränkungen; targetMarket ist 'global'.",{"category":92,"check":124,"severity":24,"summary":125},"Laufzeitstabilität","Das Skript verwendet Standard-Python-3 und `subprocess`, was es mit gängigen POSIX-Umgebungen und Windows PowerShell kompatibel macht.",{"category":45,"check":127,"severity":24,"summary":128},"README","Die README.md-Datei ist umfassend und erklärt den Zweck des Repositoriums, die Installation, die Skills und die Verwendung.",{"category":33,"check":130,"severity":42,"summary":131},"Größe der Tool-Oberfläche","Dies ist ein Skill mit einem einzigen Zweck, keine Sammlung mehrerer Tools.",{"category":40,"check":133,"severity":42,"summary":134},"Überlappende fast-synonyme Tools","Dieser Skill scheint ein einzelner Einstiegspunkt zu sein und stellt keine mehreren, unterschiedlichen Tools mit überlappender Funktionalität bereit.",{"category":45,"check":136,"severity":24,"summary":137},"Phantom-Funktionen","Die beschriebene Funktionalität zur Ausführung und Berichterstattung von Befehlen ist direkt im Skript `run_command.py` und der zugehörigen Dokumentation implementiert.",{"category":139,"check":140,"severity":24,"summary":141},"Installation","Installationsanleitung","Das README bietet klare `npx skills add`-Anweisungen und kopierbare Beispiele für die Installation und den Aufruf des Skills.",{"category":143,"check":144,"severity":24,"summary":145},"Fehler","Handlungsorientierte Fehlermeldungen","Fehler wie \"command not found\", Timeouts und Nicht-Null-Exit-Codes werden klar mit Kontext und Hinweisen zur Fehlerbehebung wie \"command failed before launch\" oder Timeout-Dauern gemeldet.",{"category":147,"check":148,"severity":24,"summary":149},"Ausführung","Angepinnte Abhängigkeiten","Das Python-Skript verwendet Standardbibliotheksfunktionen und `subprocess`, und der Shebang `#!/usr/bin/env python3` gibt den Interpreter an, was Klarheit über die Ausführungsumgebung schafft.",{"category":33,"check":151,"severity":42,"summary":152},"Dry-Run-Vorschau","Die Hauptfunktion des Skills besteht darin, einen Befehl auszuführen und seine Ausgabe zu berichten, nicht darin, zustandsändernde Operationen durchzuführen, die einen Dry-Run-Modus erfordern würden.",{"category":154,"check":155,"severity":24,"summary":156},"Protokoll","Idempotente Wiederholungen & Timeouts","Das Skript erzwingt einen harten Timeout und gibt bei Ablauf einen strukturierten Fehler zurück, und seine Kernoperation, einen Befehl auszuführen, ist von Natur aus wiederholbar.",{"category":117,"check":158,"severity":24,"summary":159},"Telemetry Opt-in","Das Skript erfasst lokale Ausführungsdaten und Git-Status zu Audit-Zwecken; es gibt keine Hinweise auf ausgehende Telemetrie.",{"category":40,"check":161,"severity":24,"summary":162},"Präziser Zweck","Der Zweck des Skills ist präzise definiert für die Ausführung dokumentierter Befehle und die Normalisierung von Nachweisen, mit klaren Abgrenzungen zu Training oder initialer Einrichtung.",{"category":40,"check":164,"severity":24,"summary":165},"Prägnanter Frontmatter","Der SKILL.md-Frontmatter ist prägnant und bietet eine klare Zusammenfassung des Zwecks des Skills und der wichtigsten Anwendungsfälle.",{"category":45,"check":167,"severity":24,"summary":168},"Prägnanter Body","Der SKILL.md-Body ist angemessen prägnant und umreißt Prinzipien und Verwendung, ohne übermäßige Inline-Details, und verweist für tiefergehendes Material auf andere Quellen.",{"category":170,"check":171,"severity":24,"summary":172},"Kontext","Progressive Offenlegung","Der SKILL.md verweist auf andere Markdown-Dateien wie `agent-operating-principles.md` und `reporting-policy.md` für tiefere Informationen.",{"category":170,"check":174,"severity":42,"summary":175},"Geforkte Erkundung","Dies ist kein stark explorativer Skill; er führt einen bestimmten Befehl aus und berichtet darüber, daher ist `context: fork` nicht anwendbar.",{"category":22,"check":177,"severity":24,"summary":178},"Anwendungsbeispiele","Das README bietet Beispiel-Prompts für vertrauenswürdige Reproduktion, die mit der Funktionalität von `minimal-run-and-audit` übereinstimmen.",{"category":22,"check":180,"severity":24,"summary":181},"Randfälle","Das Skript behandelt \"command not found\", Timeouts und Nicht-Null-Exit-Codes und dokumentiert diese Fehlerfälle und Wiederherstellungswege.",{"category":104,"check":183,"severity":42,"summary":184},"Tool-Fallback","Dieser Skill scheint nicht von externen Tools wie einem MCP-Server abzuhängen; er verwendet Standard-Python und `subprocess`.",{"category":92,"check":186,"severity":24,"summary":187},"Stack-Annahmen","Der Shebang `#!/usr/bin/env python3` und die Verwendung von Standardbibliotheken deuten auf breite Portabilität über POSIX-ähnliche Systeme und Windows PowerShell hin.",{"category":66,"check":189,"severity":24,"summary":190},"Anhalten bei unerwartetem Zustand","Das Skript prüft explizit auf Fehler wie \"command not found\" oder Timeouts und meldet diese, stoppt die Ausführung und berichtet über den Zustand.",{"category":92,"check":192,"severity":24,"summary":193},"Cross-Skill-Kopplung","Der Skill ist in sich geschlossen und konzentriert sich auf die Ausführung eines Befehls und die Berichterstattung von Ergebnissen, ohne implizite Abhängigkeiten von anderen Skills.",1778692567260,"Dieser Skill führt angegebene Befehle innerhalb eines gegebenen Repositoriums aus, erfasst deren Ausgabe (stdout, stderr), Rückgabecodes und Änderungen am Dateisystem (via git status) und normalisiert diese Informationen in eine strukturierte JSON-Nutzlast. Er behandelt Fehler wie \"command not found\", Timeouts und Nicht-Null-Exit-Codes und liefert nachprüfbare Nachweise für die Forschungsreproduktion.",[197,198,199,200,201],"Vertrauenswürdiger Ausführungspfad für die Reproduktion von KI-Repos","Erfasst Befehlsausgabe, Fehler und Dateisystemänderungen","Generiert standardisierte `repro_outputs/`-Dateien","Behandelt Ausführungs-Timeouts und Nicht-Null-Exit-Codes","Unterstützt Patch-Notizen bei Änderungen an Repository-Dateien",[203,204,205,206,207],"Erstmalige Repository-Scans oder -Aufnahme","Generische Umgebungs einrichtung","Paper-Suche oder Zielauswahl","End-to-End-Orchestrierung für sich genommen","Trainingsausführung oder Zustandsverwaltung","3.0.0","4.4.0","Bereitstellung einer vertrauenswürdigen und nachprüfbaren Methode zur Ausführung und Berichterstattung von dokumentierten Befehlen in KI-Forschungsrepositorien, um sicherzustellen, dass Nachweise konsistent für die Reproduktion erfasst werden.","Alle Prüfungen bestanden, was auf einen qualitativ hochwertigen, gut dokumentierten und sicheren Skill hinweist.",100,"Ein robuster Skill für die nachprüfbare Ausführung und Berichterstattung von Befehlen in KI-Forschungsrepositorien.",[215,216,217,218,219],"reproduction","testing","reporting","code-execution","evidence-capture","verified",[222,223,224,225],"Überprüfung dokumentierter Inferenz- oder Auswertungsbefehle","Erfassung von Nachweisen aus Smoke-Tests","Normalisierung von Ausführungsergebnissen zur Nachprüfbarkeit","Berichterstattung über Änderungen an Repository-Dateien nach der Befehlsausführung",{"codeQuality":227,"collectedAt":229,"documentation":230,"maintenance":233,"security":238,"testCoverage":240},{"hasLockfile":228},false,1778692549002,{"descriptionLength":231,"readmeSize":232},477,22701,{"closedIssues90d":8,"forks":234,"hasChangelog":235,"openIssues90d":8,"pushedAt":236,"stars":237},4,true,1778347974000,75,{"hasNpmPackage":228,"license":239,"smitheryVerified":228},"MIT",{"hasCi":235,"hasTests":235},{"updatedAt":242},1778692706692,{"basePath":244,"githubOwner":245,"githubRepo":246,"locale":18,"slug":13,"type":247},"skills/minimal-run-and-audit","lllllllama","ai-paper-reproduction-skill","skill",null,{"evaluate":250,"extract":253},{"promptVersionExtension":208,"promptVersionScoring":209,"score":212,"tags":251,"targetMarket":252,"tier":220},[215,216,217,218,219],"global",{"commitSha":254},"HEAD",{"repoId":256,"translatedFrom":257},"kd7629v5mqesxwwe9w7qtfgp7d86n6re","k1718b6r9230y7ft3t8gkmsg2986my9z",{"_creationTime":259,"_id":256,"identity":260,"providers":261,"workflow":496},1778692391648.3123,{"githubOwner":245,"githubRepo":246,"sourceUrl":14},{"classify":262,"discover":490,"github":493},{"commitSha":254,"extensions":263},[264,342,372,384,404,417,430,441,451,465,478],{"basePath":265,"description":266,"displayName":267,"installMethods":268,"rationale":269,"selectedPaths":270,"source":340,"sourceLanguage":341,"type":247},"skills/ai-research-explore","Explore-lane end-to-end orchestrator for the third research scenario: the researcher has already chosen the task family, dataset, benchmark, evaluation method, and provided SOTA references, and wants candidate-only exploration on top of `current_research` with auditable repo understanding, idea gating, and governed experiments written to `explore_outputs/`. Do not use for README-first trusted reproduction, open-ended direction finding, narrow code-only or run-only exploration, passive repo analysis, or implicit experimentation.","ai-research-explore",{"claudeCode":12},"SKILL.md frontmatter at skills/ai-research-explore/SKILL.md",[271,274,277,279,281,283,285,287,290,292,294,296,298,300,302,304,306,308,310,312,314,316,318,320,322,324,326,328,330,332,334,336,338],{"path":272,"priority":273},"SKILL.md","mandatory",{"path":275,"priority":276},"references/ai-research-explore-policy.md","medium",{"path":278,"priority":276},"references/idea-evaluation-framework.md",{"path":280,"priority":276},"references/research-campaign-spec.md",{"path":282,"priority":276},"references/smoke-validation-policy.md",{"path":284,"priority":276},"references/source-mapping-policy.md",{"path":286,"priority":276},"references/sources-naming-policy.md",{"path":288,"priority":289},"scripts/lookup/__init__.py","low",{"path":291,"priority":289},"scripts/lookup/cache_store.py",{"path":293,"priority":289},"scripts/lookup/inventory_writer.py",{"path":295,"priority":289},"scripts/lookup/normalizers.py",{"path":297,"priority":289},"scripts/lookup/providers/__init__.py",{"path":299,"priority":289},"scripts/lookup/providers/arxiv_provider.py",{"path":301,"priority":289},"scripts/lookup/providers/base.py",{"path":303,"priority":289},"scripts/lookup/providers/doi_provider.py",{"path":305,"priority":289},"scripts/lookup/providers/github_provider.py",{"path":307,"priority":289},"scripts/lookup/providers/optional_provider.py",{"path":309,"priority":289},"scripts/lookup/providers/url_provider.py",{"path":311,"priority":289},"scripts/lookup/record_schema.py",{"path":313,"priority":289},"scripts/lookup/repo_extractors.py",{"path":315,"priority":289},"scripts/lookup/source_support.py",{"path":317,"priority":289},"scripts/orchestrate_explore.py",{"path":319,"priority":289},"scripts/passes/__init__.py",{"path":321,"priority":289},"scripts/passes/atomic_idea_decomposition.py",{"path":323,"priority":289},"scripts/passes/candidate_idea_generation.py",{"path":325,"priority":289},"scripts/passes/execution_feasibility.py",{"path":327,"priority":289},"scripts/passes/idea_cards.py",{"path":329,"priority":289},"scripts/passes/idea_ranking.py",{"path":331,"priority":289},"scripts/passes/implementation_fidelity.py",{"path":333,"priority":289},"scripts/passes/improvement_bank.py",{"path":335,"priority":289},"scripts/passes/lookup_sources.py",{"path":337,"priority":289},"scripts/passes/source_mapping.py",{"path":339,"priority":289},"scripts/write_outputs.py","rule","en",{"basePath":343,"description":344,"displayName":345,"installMethods":346,"rationale":347,"selectedPaths":348,"source":340,"sourceLanguage":341,"type":247},"skills/ai-research-reproduction","Main orchestrator for README-first AI repo reproduction. Use when the user wants an end-to-end, minimal-trustworthy reproduction flow that reads the repository first, selects the smallest documented inference or evaluation target, coordinates intake, setup, trusted execution, optional trusted training, optional repository analysis, and optional paper-gap resolution, enforces conservative patch rules, records evidence assumptions deviations and human decision points, and writes the standardized `repro_outputs/` bundle. Do not use for paper summary, generic environment setup, isolated repo scanning, standalone command execution, silent protocol changes, or broad research assistance outside repository-grounded reproduction.","ai-research-reproduction",{"claudeCode":12},"SKILL.md frontmatter at skills/ai-research-reproduction/SKILL.md",[349,350,352,354,356,358,360,362,364,366,368,370],{"path":272,"priority":273},{"path":351,"priority":289},"assets/COMMANDS.template.md",{"path":353,"priority":289},"assets/LOG.template.md",{"path":355,"priority":289},"assets/PATCHES.template.md",{"path":357,"priority":289},"assets/SUMMARY.template.md",{"path":359,"priority":289},"assets/status.template.json",{"path":361,"priority":276},"references/architecture.md",{"path":363,"priority":276},"references/language-policy.md",{"path":365,"priority":276},"references/output-spec.md",{"path":367,"priority":276},"references/patch-policy.md",{"path":369,"priority":276},"references/research-safety-principles.md",{"path":371,"priority":289},"scripts/orchestrate_repro.py",{"basePath":373,"description":374,"displayName":375,"installMethods":376,"rationale":377,"selectedPaths":378,"source":340,"sourceLanguage":341,"type":247},"skills/analyze-project","Trusted-lane analysis skill for deep learning research repositories. Use when the user wants to read and understand a repository, inspect model structure and training or inference entrypoints, review configs and insertion points, or flag suspicious implementation patterns without modifying code or running heavy jobs. Do not use for active command execution, broad refactoring, speculative code adaptation, or automatic bug fixing.","analyze-project",{"claudeCode":12},"SKILL.md frontmatter at skills/analyze-project/SKILL.md",[379,380,382],{"path":272,"priority":273},{"path":381,"priority":276},"references/analysis-policy.md",{"path":383,"priority":289},"scripts/analyze_project.py",{"basePath":385,"description":386,"displayName":387,"installMethods":388,"rationale":389,"selectedPaths":390,"source":340,"sourceLanguage":341,"type":247},"skills/env-and-assets-bootstrap","Environment and assets sub-skill for README-first AI repo reproduction. Use when the task is specifically to prepare a conservative conda-first environment, checkpoint and dataset path assumptions, cache location hints, and setup notes before any run on a README-documented repository. Do not use for repo scanning, full orchestration, paper interpretation, final run reporting, or generic environment setup that is not tied to a specific reproduction target.","env-and-assets-bootstrap",{"claudeCode":12},"SKILL.md frontmatter at skills/env-and-assets-bootstrap/SKILL.md",[391,392,394,396,398,400,402],{"path":272,"priority":273},{"path":393,"priority":276},"references/assets-policy.md",{"path":395,"priority":276},"references/env-policy.md",{"path":397,"priority":289},"scripts/bootstrap_env.py",{"path":399,"priority":289},"scripts/bootstrap_env.sh",{"path":401,"priority":289},"scripts/plan_setup.py",{"path":403,"priority":289},"scripts/prepare_assets.py",{"basePath":405,"description":406,"displayName":407,"installMethods":408,"rationale":409,"selectedPaths":410,"source":340,"sourceLanguage":341,"type":247},"skills/explore-code","Explore-lane code adaptation skill for deep learning research repositories. Use when the researcher explicitly authorizes exploratory work on an isolated branch or worktree to transplant modules, adapt a backbone, add LoRA or adapter layers, replace a head, or stitch together low-risk migration ideas with summary-only records in `explore_outputs/`. Do not use for end-to-end exploration orchestration on top of `current_research`, trusted baseline reproduction, conservative debugging, environment setup, or default repository analysis.","explore-code",{"claudeCode":12},"SKILL.md frontmatter at skills/explore-code/SKILL.md",[411,412,414,416],{"path":272,"priority":273},{"path":413,"priority":276},"references/explore-policy.md",{"path":415,"priority":289},"scripts/plan_code_changes.py",{"path":339,"priority":289},{"basePath":418,"description":419,"displayName":420,"installMethods":421,"rationale":422,"selectedPaths":423,"source":340,"sourceLanguage":341,"type":247},"skills/explore-run","Explore-lane experimental execution skill for deep learning research repositories. Use when the researcher explicitly authorizes exploratory runs such as small-subset validation, short-cycle guess-and-check, batch sweeps, idle-GPU search, or quick transfer-learning trials, with results summarized in `explore_outputs/`. Do not use for end-to-end exploration orchestration on top of `current_research`, trusted baseline execution, conservative training verification, default routing, or implicit experimentation.","explore-run",{"claudeCode":12},"SKILL.md frontmatter at skills/explore-run/SKILL.md",[424,425,427,429],{"path":272,"priority":273},{"path":426,"priority":276},"references/execution-policy.md",{"path":428,"priority":289},"scripts/plan_variants.py",{"path":339,"priority":289},{"basePath":244,"description":431,"displayName":13,"installMethods":432,"rationale":433,"selectedPaths":434,"source":340,"sourceLanguage":341,"type":247},"Trusted-lane execution and reporting skill for README-first AI repo reproduction. Use when the task is specifically to capture or normalize evidence from the selected smoke test or documented inference or evaluation command and write standardized `repro_outputs/` files, including patch notes when repository files changed. Do not use for training execution, initial repo intake, generic environment setup, paper lookup, target selection, or end-to-end orchestration by itself.",{"claudeCode":12},"SKILL.md frontmatter at skills/minimal-run-and-audit/SKILL.md",[435,436,438,440],{"path":272,"priority":273},{"path":437,"priority":276},"references/reporting-policy.md",{"path":439,"priority":289},"scripts/run_command.py",{"path":339,"priority":289},{"basePath":442,"description":443,"displayName":444,"installMethods":445,"rationale":446,"selectedPaths":447,"source":340,"sourceLanguage":341,"type":247},"skills/paper-context-resolver","Optional narrow helper skill for README-first AI repo reproduction. Use only when the README and repository files leave a narrow reproduction-critical gap and the task is to resolve a specific paper detail such as dataset split, preprocessing, evaluation protocol, checkpoint mapping, or runtime assumption from primary paper sources while recording conflicts. Do not use for general paper summary, repo scanning, environment setup, command execution, title-only paper lookup, or replacing README guidance by default.","paper-context-resolver",{"claudeCode":12},"SKILL.md frontmatter at skills/paper-context-resolver/SKILL.md",[448,449],{"path":272,"priority":273},{"path":450,"priority":276},"references/paper-assisted-reproduction.md",{"basePath":452,"description":453,"displayName":454,"installMethods":455,"rationale":456,"selectedPaths":457,"source":340,"sourceLanguage":341,"type":247},"skills/repo-intake-and-plan","Narrow helper skill for README-first AI repo reproduction. Use when the task is specifically to scan a repository, read the README and common project files, extract documented commands, classify inference, evaluation, and training candidates, and return the smallest trustworthy reproduction plan to the main orchestrator. Do not use for environment setup, asset download, command execution, final reporting, paper lookup, or end-to-end orchestration.","repo-intake-and-plan",{"claudeCode":12},"SKILL.md frontmatter at skills/repo-intake-and-plan/SKILL.md",[458,459,461,463],{"path":272,"priority":273},{"path":460,"priority":276},"references/repo-scan-rules.md",{"path":462,"priority":289},"scripts/extract_commands.py",{"path":464,"priority":289},"scripts/scan_repo.py",{"basePath":466,"description":467,"displayName":468,"installMethods":469,"rationale":470,"selectedPaths":471,"source":340,"sourceLanguage":341,"type":247},"skills/run-train","Trusted-lane training execution skill for deep learning research repositories. Use when a documented or selected training command should be run conservatively for startup verification, short-run verification, full kickoff, or resume, with status, checkpoint, and metric capture written to standardized `train_outputs/`. Do not use for environment setup, exploratory sweeps, speculative idea implementation, or end-to-end orchestration.","run-train",{"claudeCode":12},"SKILL.md frontmatter at skills/run-train/SKILL.md",[472,473,475,477],{"path":272,"priority":273},{"path":474,"priority":276},"references/training-policy.md",{"path":476,"priority":289},"scripts/run_training.py",{"path":339,"priority":289},{"basePath":479,"description":480,"displayName":481,"installMethods":482,"rationale":483,"selectedPaths":484,"source":340,"sourceLanguage":341,"type":247},"skills/safe-debug","Trusted-lane debug skill for deep learning research work. Use when the user pastes a traceback, terminal error, CUDA OOM, checkpoint load failure, shape mismatch, NaN loss symptom, or training failure and wants conservative diagnosis before any patching. Do not use for broad refactoring, speculative adaptation, automatic exploratory patching, or general repository familiarization.","safe-debug",{"claudeCode":12},"SKILL.md frontmatter at skills/safe-debug/SKILL.md",[485,486,488],{"path":272,"priority":273},{"path":487,"priority":276},"references/debug-policy.md",{"path":489,"priority":289},"scripts/safe_debug.py",{"sources":491},[492],"manual",{"closedIssues90d":8,"description":494,"forks":234,"license":239,"openIssues90d":8,"pushedAt":236,"readmeSize":232,"stars":237,"topics":495},"",[],{"classifiedAt":497,"discoverAt":498,"extractAt":499,"githubAt":499,"updatedAt":497},1778692395631,1778692391648,1778692393876,[218,219,217,215,216],{"evaluatedAt":502,"extractAt":503,"updatedAt":242},1778692567369,1778692396032,[],[506,540,570,599,626,655],{"_creationTime":507,"_id":508,"community":509,"display":510,"identity":516,"providers":521,"relations":531,"tags":535,"workflow":536},1778693808357.2327,"k17b8cgj7enb26b6ek6rfjzn1x86mh4h",{"reviewCount":8},{"description":511,"installMethods":512,"name":514,"sourceUrl":515},"Aktualisiert den Context-Mode von GitHub und behebt Hooks/Einstellungen.\nZieht die neueste Version, baut sie, installiert sie, aktualisiert npm global, konfiguriert Hooks.\nTrigger: /context-mode:ctx-upgrade\n",{"claudeCode":513},"mksglu/context-mode","Context Mode","https://github.com/mksglu/context-mode",{"basePath":517,"githubOwner":518,"githubRepo":519,"locale":18,"slug":520,"type":247},"skills/ctx-upgrade","mksglu","context-mode","ctx-upgrade",{"evaluate":522,"extract":529},{"promptVersionExtension":208,"promptVersionScoring":209,"score":212,"tags":523,"targetMarket":252,"tier":220},[524,525,218,526,527,528],"context-management","llm-ops","session-continuity","productivity","mcp",{"commitSha":254,"license":530},"NOASSERTION",{"parentExtensionId":532,"repoId":533,"translatedFrom":534},"k17ezy748es7sfnbnp9phht43h86m53y","kd764b2fctbqg4b8j8y6xvmkvs86m29m","k17fqs996gpd2bggec9k1qbbns86nh4g",[218,524,525,528,527,526],{"evaluatedAt":537,"extractAt":538,"updatedAt":539},1778693713738,1778693511416,1778693808357,{"_creationTime":541,"_id":542,"community":543,"display":544,"identity":550,"providers":555,"relations":564,"tags":566,"workflow":567},1778696691708.3035,"k17br1j5s86ae90zqeyd7zcg2586mkwr",{"reviewCount":8},{"description":545,"installMethods":546,"name":548,"sourceUrl":549},"Comprehensive performance analysis, bottleneck detection, and optimization recommendations for Claude Flow swarms\n",{"claudeCode":547},"ruvnet/ruflo","Performance Analysis","https://github.com/ruvnet/ruflo",{"basePath":551,"githubOwner":552,"githubRepo":553,"locale":341,"slug":554,"type":247},".claude/skills/performance-analysis","ruvnet","ruflo","performance-analysis",{"evaluate":556,"extract":563},{"promptVersionExtension":208,"promptVersionScoring":209,"score":212,"tags":557,"targetMarket":252,"tier":220},[558,559,560,561,562,217],"performance","analysis","optimization","claude-flow","bottleneck-detection",{"commitSha":254,"license":239},{"repoId":565},"kd7ed28gj8n0y3msk5dzrp05zs86nqtc",[559,562,561,560,558,217],{"evaluatedAt":568,"extractAt":569,"updatedAt":568},1778699217174,1778696691708,{"_creationTime":571,"_id":572,"community":573,"display":574,"identity":580,"providers":584,"relations":592,"tags":595,"workflow":596},1778698144006.2202,"k172517ana4f5vj79mb22xzwsx86mksv",{"reviewCount":8},{"description":575,"installMethods":576,"name":578,"sourceUrl":579},"Audit and consolidate HubSpot reporting dashboards. Identifies unused, duplicate, or outdated dashboards. Must be performed manually — no dashboard API is available.",{"claudeCode":577},"TomGranot/hubspot-admin-skills","cleanup-dashboards","https://github.com/TomGranot/hubspot-admin-skills",{"basePath":581,"githubOwner":582,"githubRepo":583,"locale":341,"slug":578,"type":247},"skills/cleanup-dashboards","TomGranot","hubspot-admin-skills",{"evaluate":585,"extract":591},{"promptVersionExtension":208,"promptVersionScoring":209,"score":212,"tags":586,"targetMarket":252,"tier":220},[587,588,589,217,590],"hubspot","crm","maintenance","cleanup",{"commitSha":254},{"parentExtensionId":593,"repoId":594},"k17c3p8t0thc73pbc8egtz31y986mwr0","kd75kpec7arn6z2wz641vfaj8n86nab6",[590,588,587,589,217],{"evaluatedAt":597,"extractAt":598,"updatedAt":597},1778698268281,1778698144006,{"_creationTime":600,"_id":601,"community":602,"display":603,"identity":609,"providers":613,"relations":619,"tags":622,"workflow":623},1778694480889.9524,"k17cem4hc58gq77dezte6rz8mx86nkpf",{"reviewCount":8},{"description":604,"installMethods":605,"name":607,"sourceUrl":608},"Display the current state of the FPF knowledge base",{"claudeCode":606},"NeoLabHQ/context-engineering-kit","status","https://github.com/NeoLabHQ/context-engineering-kit",{"basePath":610,"githubOwner":611,"githubRepo":612,"locale":341,"slug":607,"type":247},"plugins/fpf/skills/status","NeoLabHQ","context-engineering-kit",{"evaluate":614,"extract":618},{"promptVersionExtension":208,"promptVersionScoring":209,"score":212,"tags":615,"targetMarket":252,"tier":220},[616,607,217,617],"knowledge-base","fpf",{"commitSha":254},{"parentExtensionId":620,"repoId":621},"k170dd9j7raacsjs3ta67k8cw986m50s","kd7a3rj13ezgx1wgm0jfh08hsx86n0sz",[617,616,217,607],{"evaluatedAt":624,"extractAt":625,"updatedAt":624},1778695034738,1778694480890,{"_creationTime":627,"_id":628,"community":629,"display":630,"identity":636,"providers":640,"relations":648,"tags":651,"workflow":652},1778692726926.7627,"k17dhmskz6t7wpxvd9ygy7fvsh86n695",{"reviewCount":8},{"description":631,"installMethods":632,"name":634,"sourceUrl":635},"End-of-quarter strategic review in narrative style with a bets scorecard. Use when someone says \"quarter review\", \"strategic review\", \"what happened last quarter\", \"quarterly retro\", \"bets scorecard\", \"review our bets\", \"end of quarter report\".\n",{"claudeCode":633},"marfoerst/the-pragmatic-pm","pm-strategic-review","https://github.com/marfoerst/the-pragmatic-pm",{"basePath":637,"githubOwner":638,"githubRepo":639,"locale":341,"slug":634,"type":247},"skills/pm-strategic-review","marfoerst","the-pragmatic-pm",{"evaluate":641,"extract":647},{"promptVersionExtension":208,"promptVersionScoring":209,"score":212,"tags":642,"targetMarket":252,"tier":220},[643,644,217,645,646],"product-management","strategy","review","scorecard",{"commitSha":254},{"parentExtensionId":649,"repoId":650},"k17ehawghqbe3ff7rxmq9cq1xs86nm21","kd731k864fr1ezp8r85ecbhz9986mzz7",[643,217,645,646,644],{"evaluatedAt":653,"extractAt":654,"updatedAt":653},1778693621016,1778692726926,{"_creationTime":656,"_id":657,"community":658,"display":659,"identity":665,"providers":669,"relations":678,"tags":681,"workflow":682},1778692306427.1023,"k17f0vqhj9x3ee4773kq2m8fph86n5ct",{"reviewCount":8},{"description":660,"installMethods":661,"name":663,"sourceUrl":664},"Revenue and costs tracker. AWS spend via aws ce, credits tracker, project revenue stages. Shows burn rate, runway estimate, credits expiring.",{"claudeCode":662},"Lifecycle-Innovations-Limited/claude-ops","ops-revenue","https://github.com/Lifecycle-Innovations-Limited/claude-ops",{"basePath":666,"githubOwner":667,"githubRepo":668,"locale":341,"slug":663,"type":247},"claude-ops/skills/ops-revenue","Lifecycle-Innovations-Limited","claude-ops",{"evaluate":670,"extract":677},{"promptVersionExtension":208,"promptVersionScoring":209,"score":212,"tags":671,"targetMarket":252,"tier":220},[672,673,674,675,217,676],"finance","aws","cost-tracking","revenue","dashboard",{"commitSha":254},{"parentExtensionId":679,"repoId":680},"k17d0t6ns7y6t377pfprg128hd86nm89","kd7d52tcek2e34r805zs06b10d86n39v",[673,674,676,672,217,675],{"evaluatedAt":683,"extractAt":684,"updatedAt":683},1778692873720,1778692306427]