[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"extension-plugin-Anjos2-recursive-research-en":3,"guides-for-Anjos2-recursive-research":383,"similar-k178rg6vs234c409g28ca6arbd86n5a1-en":384},{"_creationTime":4,"_id":5,"children":6,"community":45,"display":46,"evaluation":50,"identity":298,"isFallback":284,"parentExtension":301,"providers":327,"relations":332,"repo":333,"tags":380,"workflow":381},1778675291492.6482,"k178rg6vs234c409g28ca6arbd86n5a1",[7],{"_creationTime":8,"_id":9,"community":10,"display":12,"identity":18,"providers":23,"relations":38,"tags":40,"workflow":41},1778675291492.6484,"k17agpm4ma1bjqydbc6e6w2sr186m0k8",{"reviewCount":11},0,{"description":13,"installMethods":14,"name":16,"sourceUrl":17},"Investigación recursiva profunda con loop auto-regulado hasta nivel PhD. Aplicable a cualquier dominio (ciencia, tecnología, negocio, arte, humanidades). Usa WDM + Inversión Munger para decisiones autónomas, tiering de fuentes confiables, y checkpointing a disco para sobrevivir límites de contexto.",{"claudeCode":15},"Anjos2/recursive-research","recursive-research","https://github.com/Anjos2/recursive-research",{"basePath":19,"githubOwner":20,"githubRepo":16,"locale":21,"slug":16,"type":22},"plugins/recursive-research/skills/recursive-research","Anjos2","es","skill",{"evaluate":24,"extract":36},{"promptVersionExtension":25,"promptVersionScoring":26,"score":27,"tags":28,"targetMarket":34,"tier":35},"3.0.0","4.4.0",100,[29,30,31,32,33],"research","knowledge-management","llm","decision-making","documentation","global","verified",{"commitSha":37},"HEAD",{"parentExtensionId":5,"repoId":39},"kd753d1f20n4nwaapq6yp8vhd186nzxp",[32,33,30,31,29],{"evaluatedAt":42,"extractAt":43,"updatedAt":44},1778675358883,1778675291492,1778675391687,{"reviewCount":11},{"description":47,"installMethods":48,"name":49,"sourceUrl":17},"Recursive research up to PhD level across any domain (science, tech, business, arts, humanities). Source tiering, WDM + Munger inversion for autonomous decisions, disk checkpointing to survive context compaction.",{"claudeCode":16},"Recursive Research",{"_creationTime":51,"_id":52,"extensionId":5,"locale":53,"result":54,"trustSignals":282,"workflow":296},1778675336950.1025,"kn7774803zyd7drw5s91922fqh86n4e5","en",{"checks":55,"evaluatedAt":241,"extensionSummary":242,"features":243,"nonGoals":249,"practices":253,"prerequisites":258,"promptVersionExtension":25,"promptVersionScoring":26,"purpose":261,"rationale":262,"score":263,"summary":264,"tags":265,"targetMarket":34,"tier":35,"useCases":269,"workflow":274},[56,61,64,67,71,74,78,83,86,89,93,97,101,105,108,111,114,117,120,123,127,131,135,139,143,146,149,152,156,159,162,165,168,171,174,178,182,186,189,193,196,199,202,205,208,211,214,217,220,223,227,230,233,237],{"category":57,"check":58,"severity":59,"summary":60},"Practical Utility","Problem relevance","pass","The description clearly states the problem of performing recursive, in-depth research across any domain, up to PhD level.",{"category":57,"check":62,"severity":59,"summary":63},"Unique selling proposition","The extension offers significant value beyond a simple prompt by implementing advanced features like source tiering, WDM + Munger inversion for autonomous decisions, and disk checkpointing, which go beyond default LLM capabilities.",{"category":57,"check":65,"severity":59,"summary":66},"Production readiness","The plugin appears production-ready, covering the full lifecycle of a research task from initial prompting to checkpointing and iterative refinement, with clear usage examples.",{"category":68,"check":69,"severity":59,"summary":70},"Scope","Single responsibility principle","The plugin focuses on recursive research and knowledge consolidation, with a coherent set of tools and processes related to this core function.",{"category":68,"check":72,"severity":59,"summary":73},"Description quality","The displayed description accurately and concisely reflects the plugin's capabilities as detailed in the README and skill instructions.",{"category":75,"check":76,"severity":59,"summary":77},"Invocation","Scoped tools","The plugin relies on integrated Claude Code tools like WebSearch and WebFetch, and recommends specific MCPs like Firecrawl, which are generally scoped rather than arbitrary shell execution.",{"category":79,"check":80,"severity":81,"summary":82},"Documentation","Configuration & parameter reference","info","While the skill provides a comprehensive workflow, explicit documentation on the precedence order of configuration files or environment variables beyond the initial prompt interaction is not detailed.",{"category":68,"check":84,"severity":59,"summary":85},"Tool naming","The primary tool is invoked as `/recursive-research:recursive-research`, which is descriptive and specific to the plugin's function.",{"category":68,"check":87,"severity":59,"summary":88},"Minimal I/O surface","The skill prompts for specific inputs and generates structured output files, avoiding overly broad data requests or diagnostic dumps.",{"category":90,"check":91,"severity":59,"summary":92},"License","License usability","The extension is licensed under MIT and declares it via a LICENSE file and manifest.",{"category":94,"check":95,"severity":59,"summary":96},"Maintenance","Commit recency","The last commit was on April 22, 2026, which is within the last 90 days.",{"category":94,"check":98,"severity":99,"summary":100},"Dependency Management","not_applicable","The extension does not appear to bundle or directly manage third-party dependencies beyond Claude Code's integrated tools and recommended MCPs.",{"category":102,"check":103,"severity":99,"summary":104},"Security","Secret Management","The extension does not appear to handle or expose any secrets.",{"category":102,"check":106,"severity":59,"summary":107},"Injection","The skill's process for handling user input and external data seems robust, with clear instructions on treating sources as data and no indication of executing arbitrary instructions from loaded content.",{"category":102,"check":109,"severity":59,"summary":110},"Transitive Supply-Chain Grenades","All necessary components appear to be bundled within the repository, and there's no indication of runtime fetching of code or instructions from external URLs.",{"category":102,"check":112,"severity":59,"summary":113},"Sandbox Isolation","The plugin operates within the project directory for output and relies on standard Claude Code tools, with no indication of attempting to modify files outside its designated scope.",{"category":102,"check":115,"severity":59,"summary":116},"Sandbox escape primitives","No detached-process spawns or retry loops around denied tool calls were detected in the provided source.",{"category":102,"check":118,"severity":59,"summary":119},"Data Exfiltration","The plugin emphasizes local execution and privacy, with no indication of submitting confidential data to third parties.",{"category":102,"check":121,"severity":59,"summary":122},"Hidden Text Tricks","The bundled content and descriptions appear free of hidden steering tricks, invisible characters, or other obfuscation methods.",{"category":124,"check":125,"severity":59,"summary":126},"Hooks","Opaque code execution","The skill logic is primarily defined in readable markdown and relies on Claude Code's built-in capabilities, with no obfuscated or dynamically fetched code.",{"category":128,"check":129,"severity":59,"summary":130},"Portability","Structural Assumption","The skill explicitly handles the creation of the 'memoria/' directory if it doesn't exist and uses relative paths, demonstrating good portability.",{"category":132,"check":133,"severity":59,"summary":134},"Trust","Issues Attention","There are 0 open and 0 closed issues in the last 90 days, indicating a new or inactive but not neglected state.",{"category":136,"check":137,"severity":59,"summary":138},"Versioning","Release Management","The plugin has a clear version (2.2.0) declared in the README, plugin.json, and SKILL.md.",{"category":140,"check":141,"severity":81,"summary":142},"Code Execution","Validation","The skill interactively prompts for inputs, but explicit schema-based validation of all arguments, especially file paths and URLs, is not detailed.",{"category":102,"check":144,"severity":59,"summary":145},"Unguarded Destructive Operations","The primary operations involve file I/O for checkpoints and research data, which are managed locally and not considered destructive in the context of system modification.",{"category":140,"check":147,"severity":59,"summary":148},"Error Handling","The skill is designed to provide clear user feedback and error messages, especially during the initial setup and research process.",{"category":140,"check":150,"severity":59,"summary":151},"Logging","The plugin explicitly creates detailed log files for research checkpoints and outcomes within the project's 'memoria/' directory, providing an audit trail.",{"category":153,"check":154,"severity":59,"summary":155},"Compliance","GDPR","The extension clearly states it does not collect user data, ensuring compliance with GDPR.",{"category":153,"check":157,"severity":59,"summary":158},"Target market","The extension's functionality is global and not tied to any specific geographic or legal jurisdiction.",{"category":128,"check":160,"severity":59,"summary":161},"Runtime stability","The skill relies on standard markdown and Claude Code's built-in tools, with no apparent assumptions about specific operating systems or shells.",{"category":79,"check":163,"severity":59,"summary":164},"README","The README is comprehensive, clearly stating the plugin's purpose, features, and installation/usage instructions.",{"category":68,"check":166,"severity":99,"summary":167},"Tool surface size","This is a single-tool extension, so the tool surface size check is not applicable.",{"category":75,"check":169,"severity":99,"summary":170},"Overlapping near-synonym tools","As a single-tool extension, there are no overlapping tools to evaluate.",{"category":79,"check":172,"severity":59,"summary":173},"Phantom features","All features mentioned in the README, such as WDM+Munger inversion and disk checkpointing, are implemented and described in the SKILL.md.",{"category":175,"check":176,"severity":59,"summary":177},"Install","Installation instruction","Installation instructions are clear, offering multiple options and including copy-pasteable commands and examples.",{"category":179,"check":180,"severity":59,"summary":181},"Errors","Actionable error messages","The skill is designed to provide clear user feedback and error messages, including guidance on resuming research or creating necessary directories.",{"category":183,"check":184,"severity":99,"summary":185},"Execution","Pinned dependencies","The extension does not rely on third-party dependencies that require pinning via lockfiles.",{"category":68,"check":187,"severity":99,"summary":188},"Dry-run preview","The extension is primarily analytical and does not perform destructive operations or send data outward, making a dry-run feature not applicable.",{"category":190,"check":191,"severity":99,"summary":192},"Protocol","Idempotent retry & timeouts","The extension's operations are largely local file I/O and tool calls within Claude Code, which handle their own timeouts and retries; no external mutating operations require this.",{"category":153,"check":194,"severity":59,"summary":195},"Telemetry opt-in","The plugin explicitly states it does not collect any user data, meaning telemetry is effectively off by default and there's nothing to opt into.",{"category":75,"check":197,"severity":99,"summary":198},"Name collisions","The repository contains a single plugin and skill, so there are no name collisions to evaluate.",{"category":75,"check":200,"severity":99,"summary":201},"Hooks-off mechanism","The plugin does not appear to utilize hooks that would require a hooks-off mechanism.",{"category":75,"check":203,"severity":99,"summary":204},"Hook matcher tightness","No hooks are declared by this plugin.",{"category":102,"check":206,"severity":99,"summary":207},"Hook security","The plugin does not employ any hooks.",{"category":124,"check":209,"severity":99,"summary":210},"Silent prompt rewriting","The plugin does not have a UserPromptSubmit hook.",{"category":102,"check":212,"severity":99,"summary":213},"Permission Hook","There are no PermissionRequest hooks implemented in this plugin.",{"category":153,"check":215,"severity":99,"summary":216},"Hook privacy","The plugin does not utilize hooks that involve logging or telemetry data transmission.",{"category":140,"check":218,"severity":99,"summary":219},"Hook dependency","No hooks are present in the plugin.",{"category":79,"check":221,"severity":59,"summary":222},"Feature Transparency","Critical functionalities like WDM+Munger and checkpointing are explained in the README and SKILL.md. No hooks are declared in plugin.json.",{"category":224,"check":225,"severity":59,"summary":226},"Convention","Layout convention adherence","The plugin follows the expected layout with .claude-plugin/marketplace.json and plugin.json, and the skill is in the correct path. No bin/ directory is present.",{"category":224,"check":228,"severity":59,"summary":229},"Plugin state","The plugin correctly manages its state within the project directory ('memoria/'), which aligns with standard practices for persistence and uninstall.",{"category":102,"check":231,"severity":99,"summary":232},"Keychain-stored secrets","The plugin does not handle secrets that would require keychain storage.",{"category":234,"check":235,"severity":59,"summary":236},"Dependencies","Tagged release sourcing","The plugin directly bundles its skill and relies on Claude Code's built-in tools and recommended MCPs, not external MCP servers sourced from potentially unstable references.",{"category":238,"check":239,"severity":59,"summary":240},"Installation","Clean uninstall","The plugin operates locally and creates files within the project directory; uninstalling Claude Code or removing the plugin should not leave persistent background processes or system modifications.",1778675336670,"This plugin provides a recursive research capability that iterates through topics, sources, and decision frameworks to achieve PhD-level understanding. It supports various domains, tiers sources, survives context loss via disk checkpoints, and is guided by WDM and Munger inversion principles. It can resume paused research and provides detailed output files.",[244,245,246,247,248],"Recursive research up to PhD level","Multi-domain applicability (science, tech, business, arts, humanities)","Source tiering and automatic rejection of unreliable sources","Autonomous decision-making via WDM and Munger inversion","Disk checkpointing to survive context compaction and enable resuming research",[250,251,252],"Performing arbitrary web scraping without a research focus","Replacing structured academic databases for highly specialized queries","Generating novel scientific hypotheses without user seed input",[254,255,256,257],"Source tiering","Autonomous decision-making","Knowledge consolidation","Error handling",[259,260],"Claude Code environment","Local project directory for output files","To empower users to conduct deep, autonomous research across any subject matter, generating comprehensive and well-supported knowledge bases.","Excellent implementation of advanced research capabilities with strong documentation and adherence to best practices. Minor info finding on explicit parameter documentation.",95,"A highly capable plugin for in-depth, recursive research across any domain, employing advanced AI techniques.",[29,30,266,267,268],"ai-agent","information-retrieval","claude-code",[270,271,272,273],"Deeply understanding a new field for informed decision-making","Preparing for technical documentation, academic papers, or proposals","Identifying the state-of-the-art and knowledge gaps in a domain","Conducting comprehensive literature reviews",[275,276,277,278,279,280,281],"User provides research seed, mode, local paths, and preferences.","Skill identifies initial research threads using WDM and Munger inversion.","Skill detects and prioritizes available MCPs and native tools.","Skill iterates through cycles, selecting threads, investigating sources, and consolidating findings.","Sources are tiered, and disk checkpoints are saved each cycle.","Research closes when PhD criteria are met, cycle cap is reached, or user pauses.","Final synthesis, action items, and knowledge gaps are generated.",{"codeQuality":283,"collectedAt":285,"documentation":286,"maintenance":289,"security":293,"testCoverage":295},{"hasLockfile":284},false,1778675309767,{"descriptionLength":287,"readmeSize":288},212,10252,{"closedIssues90d":11,"forks":11,"hasChangelog":284,"manifestVersion":290,"openIssues90d":11,"pushedAt":291,"stars":292},"2.2.0",1776902259000,7,{"hasNpmPackage":284,"license":294,"smitheryVerified":284},"MIT",{"hasCi":284,"hasTests":284},{"updatedAt":297},1778675336950,{"basePath":299,"githubOwner":20,"githubRepo":16,"locale":53,"slug":16,"type":300},"plugins/recursive-research","plugin",{"_creationTime":302,"_id":303,"community":304,"display":305,"identity":308,"parentExtension":311,"providers":312,"relations":322,"tags":323,"workflow":324},1778675291492.648,"k171vfst8y7c4bkw07sx66bgb986n8e0",{"reviewCount":11},{"description":306,"installMethods":307,"name":49,"sourceUrl":17},"Recursive research skill/plugin for Claude Code by Joseph Huayhualla (@Anjos2)",{"claudeCode":15},{"basePath":309,"githubOwner":20,"githubRepo":16,"locale":53,"slug":16,"type":310},"","marketplace",null,{"evaluate":313,"extract":317},{"promptVersionExtension":314,"promptVersionScoring":26,"score":263,"tags":315,"targetMarket":34,"tier":35},"3.1.0",[266,316,30,16,268],"research-tool",{"commitSha":37,"license":294,"marketplace":318,"plugin":320},{"name":16,"pluginCount":319},1,{"mcpCount":11,"provider":321,"skillCount":11},"classify",{"repoId":39},[266,268,30,16,316],{"evaluatedAt":325,"extractAt":43,"updatedAt":326},1778675309479,1778675391455,{"evaluate":328,"extract":330},{"promptVersionExtension":25,"promptVersionScoring":26,"score":263,"tags":329,"targetMarket":34,"tier":35},[29,30,266,267,268],{"commitSha":37,"license":294,"plugin":331},{"mcpCount":11,"provider":321,"skillCount":319},{"parentExtensionId":303,"repoId":39},{"_creationTime":334,"_id":39,"identity":335,"providers":336,"workflow":376},1778675287813.5454,{"githubOwner":20,"githubRepo":16,"sourceUrl":17},{"classify":337,"discover":367,"github":370},{"commitSha":37,"extensions":338},[339,352,361],{"basePath":309,"description":306,"displayName":16,"installMethods":340,"rationale":341,"selectedPaths":342,"source":351,"sourceLanguage":53,"type":310},{"claudeCode":15},"marketplace.json at .claude-plugin/marketplace.json",[343,346,348],{"path":344,"priority":345},".claude-plugin/marketplace.json","mandatory",{"path":347,"priority":345},"README.md",{"path":349,"priority":350},"LICENSE","high","rule",{"basePath":299,"description":47,"displayName":16,"installMethods":353,"license":294,"rationale":354,"selectedPaths":355,"source":351,"sourceLanguage":53,"type":300},{"claudeCode":16},"plugin manifest at plugins/recursive-research/.claude-plugin/plugin.json",[356,358],{"path":357,"priority":345},".claude-plugin/plugin.json",{"path":359,"priority":360},"skills/recursive-research/SKILL.md","medium",{"basePath":19,"description":13,"displayName":16,"installMethods":362,"rationale":363,"selectedPaths":364,"source":351,"sourceLanguage":21,"type":22},{"claudeCode":15},"SKILL.md frontmatter at plugins/recursive-research/skills/recursive-research/SKILL.md",[365],{"path":366,"priority":345},"SKILL.md",{"sources":368},[369],"manual",{"closedIssues90d":11,"description":371,"forks":11,"license":294,"openIssues90d":11,"pushedAt":291,"readmeSize":288,"stars":292,"topics":372},"Claude Code skill for recursive research up to PhD level across any domain. Source tiering, WDM + Munger inversion for autonomous decisions, and disk checkpointing to survive context compaction.",[266,268,373,30,374,316,16,375],"claude-code-skill","mental-models","weighted-decision-matrix",{"classifiedAt":377,"discoverAt":378,"extractAt":379,"githubAt":379,"updatedAt":377},1778675291330,1778675287813,1778675289690,[266,268,267,30,29],{"evaluatedAt":297,"extractAt":43,"updatedAt":382},1778675391850,[],[385,421,450,478,508,538],{"_creationTime":386,"_id":387,"community":388,"display":389,"identity":395,"providers":398,"relations":414,"tags":417,"workflow":418},1778683100520.2961,"k1754vkdjckrkqvz9x7tjrvhzn86n1gc",{"reviewCount":11},{"description":390,"installMethods":391,"name":393,"sourceUrl":394},"AI music generation workflow for Suno - album concepts, lyrics, prompts, mastering, release",{"claudeCode":392},"bitwize-music","Claude AI Music Skills","https://github.com/bitwize-music-studio/claude-ai-music-skills",{"basePath":309,"githubOwner":396,"githubRepo":397,"locale":53,"slug":397,"type":300},"bitwize-music-studio","claude-ai-music-skills",{"evaluate":399,"extract":410},{"promptVersionExtension":25,"promptVersionScoring":26,"score":27,"tags":400,"targetMarket":34,"tier":35},[401,402,403,404,405,406,407,408,409,268],"music-generation","ai","suno","audio-production","workflow","lyrics","mastering","cli","python",{"commitSha":37,"license":411,"plugin":412},"CC0-1.0",{"mcpCount":11,"provider":321,"skillCount":413},54,{"parentExtensionId":415,"repoId":416},"k17bfryzkzywswf1bkgrtch16d86n8t9","kd70cgrajsrnk5gmq60rhq30zd86nyc0",[402,404,268,408,406,407,401,409,403,405],{"evaluatedAt":419,"extractAt":420,"updatedAt":419},1778683131031,1778683100520,{"_creationTime":422,"_id":423,"community":424,"display":425,"identity":430,"providers":432,"relations":442,"tags":445,"workflow":446},1778699316533.7866,"k17d3jtp70vmbqjhnze3n53ra586n5r8",{"reviewCount":11},{"description":426,"installMethods":427,"name":428,"sourceUrl":429},"Search academic papers via OpenAlex — find papers by keyword, look up details by DOI, with pagination and sorting",{"claudeCode":428},"paper-search","https://github.com/ykdojo/paper-search",{"basePath":309,"githubOwner":431,"githubRepo":428,"locale":53,"slug":428,"type":300},"ykdojo",{"evaluate":433,"extract":440},{"promptVersionExtension":25,"promptVersionScoring":26,"score":27,"tags":434,"targetMarket":34,"tier":35},[435,436,437,438,29,439],"academic","search","papers","openalex","citations",{"commitSha":37,"license":294,"plugin":441},{"mcpCount":11,"provider":321,"skillCount":319},{"parentExtensionId":443,"repoId":444},"k17abfkyvjasac4fgc8v24wz6186mvem","kd78zpgf1ptwq5s0gcz3yqr9n186mvy5",[435,439,438,437,29,436],{"evaluatedAt":447,"extractAt":448,"updatedAt":449},1778699343032,1778699316533,1778699386711,{"_creationTime":451,"_id":452,"community":453,"display":454,"identity":460,"providers":464,"relations":471,"tags":474,"workflow":475},1778690773482.4834,"k179sm2kkyd7r7nz9jsx62jm9x86mw4a",{"reviewCount":11},{"description":455,"installMethods":456,"name":458,"sourceUrl":459},"Look up and read Hugging Face paper pages in markdown, and use the papers API for structured metadata like authors, linked models, datasets, Spaces, and media URLs when needed.",{"claudeCode":457},"huggingface-papers","Hugging Face Papers","https://github.com/huggingface/skills",{"basePath":461,"githubOwner":462,"githubRepo":463,"locale":53,"slug":457,"type":300},"skills/huggingface-papers","huggingface","skills",{"evaluate":465,"extract":469},{"promptVersionExtension":25,"promptVersionScoring":26,"score":27,"tags":466,"targetMarket":34,"tier":35},[462,437,467,402,29,468],"arxiv","metadata",{"commitSha":37,"license":470},"Apache-2.0",{"parentExtensionId":472,"repoId":473},"k17es3r8wd37t5rrwqcpp5kwrh86mxx8","kd72xwt5xnc0ktc4p7smzfcp3986m959",[402,467,462,468,437,29],{"evaluatedAt":476,"extractAt":477,"updatedAt":476},1778690901306,1778690773482,{"_creationTime":479,"_id":480,"community":481,"display":482,"identity":487,"providers":490,"relations":501,"tags":504,"workflow":505},1778699234184.6104,"k17brg5egdw1jbncj1j4wfv3fh86n639",{"reviewCount":11},{"description":483,"installMethods":484,"name":485,"sourceUrl":486},"Multi-agent orchestration system for Claude Code",{"claudeCode":485},"oh-my-claudecode","https://github.com/Yeachan-Heo/oh-my-claudecode",{"basePath":309,"githubOwner":488,"githubRepo":485,"locale":489,"slug":485,"type":300},"Yeachan-Heo","fr",{"evaluate":491,"extract":498},{"promptVersionExtension":25,"promptVersionScoring":26,"score":492,"tags":493,"targetMarket":34,"tier":35},99,[494,495,496,266,497,405,408],"multi-agent","orchestration","automation","development-tool",{"commitSha":37,"license":294,"plugin":499},{"mcpCount":11,"provider":321,"skillCount":500},38,{"parentExtensionId":502,"repoId":503},"k17faea1fme9qa3pf4hzs9x78d86n2kj","kd74zv63fryf9prygtq7gf4es986n22y",[266,496,408,497,494,495,405],{"evaluatedAt":506,"extractAt":507,"updatedAt":506},1778699261932,1778699234184,{"_creationTime":509,"_id":510,"community":511,"display":512,"identity":518,"providers":520,"relations":530,"tags":533,"workflow":534},1778692878074.8347,"k178v80t9azwkv73nsx1a4r35n86m226",{"reviewCount":11},{"description":513,"installMethods":514,"name":516,"sourceUrl":517},"Worktrunk is a CLI for Git worktree management, designed for parallel AI agent workflows. This plugin provides configuration guidance (LLM commit messages, project hooks, worktree paths) and automatic activity tracking (🤖/💬 indicators in `wt list` showing active Claude sessions).",{"claudeCode":515},"worktrunk","Worktrunk","https://github.com/max-sixty/worktrunk",{"basePath":309,"githubOwner":519,"githubRepo":515,"locale":53,"slug":515,"type":300},"max-sixty",{"evaluate":521,"extract":526},{"promptVersionExtension":25,"promptVersionScoring":26,"score":492,"tags":522,"targetMarket":34,"tier":35},[523,524,408,266,496,525],"git","worktree","developer-tool",{"commitSha":37,"license":527,"plugin":528},"MIT OR Apache-2.0",{"mcpCount":11,"provider":321,"skillCount":529},2,{"parentExtensionId":531,"repoId":532},"k17933gw8zcjnsjg1xt0r1t2a586n11x","kd735mff6mvfxbagjk8sv9pg9s86ngdh",[266,496,408,525,523,524],{"evaluatedAt":535,"extractAt":536,"updatedAt":537},1778692927488,1778692878074,1778693214478,{"_creationTime":539,"_id":540,"community":541,"display":542,"identity":547,"providers":551,"relations":559,"tags":562,"workflow":563},1778698371743.7026,"k175pn6bnbbfyybfj0c7g5xm8h86nc75",{"reviewCount":11},{"description":543,"installMethods":544,"name":545,"sourceUrl":546},"Automatic long-term memory for Claude Code via Hindsight. Recalls relevant memories before each prompt, retains conversation transcripts, and provides knowledge page tools.",{"claudeCode":545},"hindsight-memory","https://github.com/vectorize-io/hindsight",{"basePath":548,"githubOwner":549,"githubRepo":550,"locale":53,"slug":268,"type":300},"hindsight-integrations/claude-code","vectorize-io","hindsight",{"evaluate":552,"extract":557},{"promptVersionExtension":25,"promptVersionScoring":26,"score":492,"tags":553,"targetMarket":34,"tier":35},[554,550,31,555,268,556],"memory","long-term-memory","agent",{"commitSha":37,"license":294,"plugin":558},{"mcpCount":11,"provider":321,"skillCount":319},{"parentExtensionId":560,"repoId":561},"k178tthvkav2341vt1q8d8j8yn86ncbf","kd7etqyey7pjacrkp7wzfh4vts86nq1z",[556,268,550,31,555,554],{"evaluatedAt":564,"extractAt":565,"updatedAt":564},1778698417908,1778698371743]