[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"extension-cli-vectorize-io-hindsight-api-zh-CN":3,"guides-for-vectorize-io-hindsight-api":643,"similar-k170nrrnsz42v8a6ag7fan0p4h86mcnn-zh-CN":644},{"_creationTime":4,"_id":5,"children":6,"community":7,"display":9,"evaluation":14,"identity":250,"isFallback":233,"parentExtension":253,"providers":254,"relations":259,"repo":261,"tags":640,"workflow":641},1778698371743.7056,"k170nrrnsz42v8a6ag7fan0p4h86mcnn",[],{"reviewCount":8},0,{"description":10,"installMethods":11,"name":12,"sourceUrl":13},"Hindsight: Agent Memory That  Learns",{"pypi":12},"hindsight-api","https://github.com/vectorize-io/hindsight",{"_creationTime":15,"_id":16,"extensionId":5,"locale":17,"result":18,"trustSignals":231,"workflow":248},1778698625700.7974,"kn77k539k7m97tyaks1bjr8mn586n6xy","en",{"checks":19,"evaluatedAt":198,"extensionSummary":199,"features":200,"nonGoals":206,"promptVersionExtension":210,"promptVersionScoring":211,"purpose":212,"rationale":213,"score":214,"summary":215,"tags":216,"targetMarket":224,"tier":225,"useCases":226},[20,25,28,31,35,38,42,46,49,52,56,60,63,67,70,73,76,79,82,85,89,93,97,101,106,109,113,116,120,123,126,129,132,135,138,142,146,149,152,156,159,163,166,169,172,175,178,181,184,187,191,194],{"category":21,"check":22,"severity":23,"summary":24},"Practical Utility","Problem relevance","pass","The description clearly identifies the problem of AI agents needing persistent memory and addresses it with a temporal, semantic, and entity-based architecture.",{"category":21,"check":26,"severity":23,"summary":27},"Unique selling proposition","The extension offers a multi-strategy retrieval system (TEMPR) with entity graph and temporal reasoning, going beyond basic memory storage and LLM defaults.",{"category":21,"check":29,"severity":23,"summary":30},"Production readiness","The CLI tool provides installation, server start, configuration, and Docker options, covering the full lifecycle for an agent memory system.",{"category":32,"check":33,"severity":23,"summary":34},"Scope","Single responsibility principle","The extension focuses on providing an agent memory system, with related CLI and API components, without introducing unrelated functionalities.",{"category":32,"check":36,"severity":23,"summary":37},"Description quality","The displayed description accurately reflects the extension's purpose of providing a memory system for AI agents.",{"category":39,"check":40,"severity":23,"summary":41},"Invocation","Scoped tools","The CLI commands exposed appear to be scoped verb-noun actions related to managing the hindsight API server.",{"category":43,"check":44,"severity":23,"summary":45},"Documentation","Configuration & parameter reference","All configuration options, including environment variables and their defaults, are clearly documented in the README.",{"category":32,"check":47,"severity":23,"summary":48},"Tool naming","The CLI commands (`hindsight-api`, `hindsight-local-mcp`) are descriptive and relevant to the extension's domain.",{"category":32,"check":50,"severity":23,"summary":51},"Minimal I/O surface","The CLI options appear to be specific flags for server configuration, with no indication of excessive or unnecessary input/output.",{"category":53,"check":54,"severity":23,"summary":55},"License","License usability","The project includes a MIT license file, which is a permissive open-source license.",{"category":57,"check":58,"severity":23,"summary":59},"Maintenance","Commit recency","The last commit was on 2026-05-13, indicating recent maintenance.",{"category":57,"check":61,"severity":23,"summary":62},"Dependency Management","The project uses Poetry for dependency management, indicated by pyproject.toml, which typically includes lockfiles.",{"category":64,"check":65,"severity":23,"summary":66},"Security","Secret Management","The README indicates API keys are expected via environment variables, not hardcoded or passed directly on the command line.",{"category":64,"check":68,"severity":23,"summary":69},"Injection","The extension appears to load data primarily from PostgreSQL and LLM providers, with no indication of processing untrusted external files as code.",{"category":64,"check":71,"severity":23,"summary":72},"Transitive Supply-Chain Grenades","Dependencies are managed via pip and pyproject.toml, suggesting a controlled supply chain. No runtime remote code execution is apparent.",{"category":64,"check":74,"severity":23,"summary":75},"Sandbox Isolation","The extension focuses on server operations and memory management; there's no evidence of file operations outside its scope or user-specific paths.",{"category":64,"check":77,"severity":23,"summary":78},"Sandbox escape primitives","No detached process spawns or deny-retry loops were observed in the provided script snippets.",{"category":64,"check":80,"severity":23,"summary":81},"Data Exfiltration","Secrets are handled via environment variables, and outbound calls are to LLM providers, with no indication of unauthorized data exfiltration.",{"category":64,"check":83,"severity":23,"summary":84},"Hidden Text Tricks","The README and provided code snippets do not contain any hidden text tricks or malicious Unicode characters.",{"category":86,"check":87,"severity":23,"summary":88},"Hooks","Opaque code execution","The Python scripts appear to be standard, readable Python code, with no signs of obfuscation like base64 payloads or minified bundles.",{"category":90,"check":91,"severity":23,"summary":92},"Portability","Structural Assumption","The CLI tool is primarily for server management and does not appear to make assumptions about user project structure.",{"category":94,"check":95,"severity":23,"summary":96},"Trust","Issues Attention","With 70 issues opened and 233 closed in the last 90 days, the closure rate is high (approximately 77%), indicating good maintainer engagement.",{"category":98,"check":99,"severity":23,"summary":100},"Versioning","Release Management","The `pyproject.toml` file declares a version of `0.6.1`, and there is a CHANGELOG.md, indicating a meaningful versioning system.",{"category":102,"check":103,"severity":104,"summary":105},"Execution","Validation","not_applicable","The evaluation of schema validation for input/output is not directly applicable as the CLI tool primarily serves as a server runner and configuration manager, not a data processing tool with complex argument validation.",{"category":64,"check":107,"severity":23,"summary":108},"Unguarded Destructive Operations","The CLI tool's primary function is starting a server or managing memory. There are no apparent destructive operations that would require an unguarded confirmation gate.",{"category":110,"check":111,"severity":23,"summary":112},"Code Execution","Error Handling","Python applications typically have robust error handling, and the structure of the CLI suggests it would exit with non-zero codes on failure.",{"category":110,"check":114,"severity":23,"summary":115},"Logging","The README mentions `--log-level debug` and verbose logging, indicating that logging is implemented and can be controlled.",{"category":117,"check":118,"severity":104,"summary":119},"Compliance","GDPR","The extension's core function is memory management for AI agents, and it does not appear to operate on personal data without sanitization. Secrets are managed via env vars.",{"category":117,"check":121,"severity":23,"summary":122},"Target market","The extension is designed for general AI agent use cases and does not exhibit any regional or jurisdictional limitations, thus 'global' is appropriate.",{"category":90,"check":124,"severity":23,"summary":125},"Runtime stability","The tool is written in Python and uses standard libraries, suggesting good cross-platform compatibility on POSIX-compliant systems.",{"category":43,"check":127,"severity":23,"summary":128},"README","The README is comprehensive, clearly stating the purpose, installation, quick start, CLI options, configuration, and Docker usage.",{"category":32,"check":130,"severity":104,"summary":131},"Tool surface size","The CLI exposes a few top-level commands related to running the server and related utilities, fitting well within the expected range.",{"category":39,"check":133,"severity":23,"summary":134},"Overlapping near-synonym tools","The exposed commands (`hindsight-api`, `hindsight-local-mcp`, etc.) have distinct and non-overlapping functions.",{"category":43,"check":136,"severity":23,"summary":137},"Phantom features","All features mentioned in the README, such as installation, server start, and API usage, correspond to implemented functionality.",{"category":139,"check":140,"severity":23,"summary":141},"Install","Installation instruction","The README provides clear installation instructions using pip and includes runnable invocation examples for the server and API.",{"category":143,"check":144,"severity":23,"summary":145},"Errors","Actionable error messages","Python applications typically provide actionable error messages, and the CLI's structure implies errors would be reported clearly.",{"category":102,"check":147,"severity":23,"summary":148},"Pinned dependencies","The `pyproject.toml` file and its dependency on `hindsight-api-slim` indicate pinned dependencies, likely with a lockfile mechanism.",{"category":32,"check":150,"severity":104,"summary":151},"Dry-run preview","The extension is primarily a server and memory management tool; state-changing operations like sending emails or infrastructure changes are not its focus, making a dry-run preview not applicable.",{"category":153,"check":154,"severity":104,"summary":155},"Protocol","Idempotent retry & timeouts","The extension is a local server and memory management tool; it does not appear to involve remote calls or state-changing operations that would require complex retry or timeout logic.",{"category":117,"check":157,"severity":23,"summary":158},"Telemetry opt-in","There is no mention of telemetry collection in the README or provided code snippets, suggesting it is either absent or strictly opt-in by default.",{"category":39,"check":160,"severity":161,"summary":162},"Hierarchical discoverable arguments","info","The CLI provides a few top-level commands, but it is not a deeply hierarchical structure. The `--help` flag is available for each command.",{"category":39,"check":164,"severity":23,"summary":165},"Unknown-flag handling","Python argument parsers typically provide built-in handling for unknown flags, usually with a 'did you mean' suggestion.",{"category":64,"check":167,"severity":23,"summary":168},"Secrets not in argv","The README explicitly states API keys are set via environment variables, aligning with security best practices against passing secrets directly in argv.",{"category":39,"check":170,"severity":23,"summary":171},"Parseable --version","The pyproject.toml specifies a version, and Python CLIs typically output a single-line version string upon request.",{"category":39,"check":173,"severity":23,"summary":174},"Flag precedence","The README clearly documents configuration via environment variables, and the `--help` output likely implies flag precedence over environment variables.",{"category":39,"check":176,"severity":104,"summary":177},"Schema introspection","As the CLI primarily runs a server and provides configuration, a machine-readable schema introspection command for its operations is not a typical or necessary feature.",{"category":39,"check":179,"severity":23,"summary":180},"Prompt bypass","The CLI tool's main function is server startup and configuration, which typically do not involve interactive prompts requiring bypass flags.",{"category":139,"check":182,"severity":23,"summary":183},"No unnecessary global install","The primary installation method is `pip install`, which installs the package into the current Python environment, avoiding unnecessary global installs.",{"category":139,"check":185,"severity":23,"summary":186},"Pipe-to-shell installer","The primary installation method uses `pip install`, avoiding insecure pipe-to-shell patterns.",{"category":188,"check":189,"severity":23,"summary":190},"Output","Stream separation","Python CLIs typically separate stdout for data and stderr for diagnostics, and the logging options suggest this separation is maintained.",{"category":188,"check":192,"severity":104,"summary":193},"Structured output mode","The CLI tool's main function is to start a server and manage memory, not to return data that requires a structured output mode for agent consumption.",{"category":195,"check":196,"severity":23,"summary":197},"Subprocess behavior","Signal handling","Python applications generally handle signals gracefully, ensuring clean shutdowns and propagation to child processes.",1778698624699,"Hindsight is a CLI tool and API that provides AI agents with persistent memory capabilities, leveraging PostgreSQL with pgvector for temporal, semantic, and entity-based storage and retrieval. It supports multiple LLM providers and offers programmatic access via Python SDK.",[201,202,203,204,205],"Persistent memory for AI agents","Temporal, semantic, and entity-based recall","Support for multiple LLM providers","Python SDK for programmatic access","Configurable disposition traits for opinion formation",[207,208,209],"Acting as a general-purpose database","Replacing the core LLM reasoning capabilities","Providing a front-end UI for end-users","3.0.0","4.4.0","To equip AI agents with sophisticated, persistent memory that mimics human recall, enabling better contextual understanding and reasoning over time.","The extension is highly polished with excellent documentation, clear features, and strong security practices. A minor info-level finding regarding the CLI's hierarchical argument structure prevents a perfect score.",96,"A robust and well-documented AI agent memory system with strong security and usability.",[217,218,219,220,221,222,223],"ai","memory","agent","llm","database","postgresql","cli","global","verified",[227,228,229,230],"Building AI agents that need to remember past interactions and facts","Developing conversational AI that maintains context over long periods","Creating agents that can reason about events based on their temporal order","Enhancing agent capabilities with structured knowledge representation",{"codeQuality":232,"collectedAt":234,"documentation":235,"maintenance":238,"security":244,"testCoverage":247},{"hasLockfile":233},true,1778698609904,{"descriptionLength":236,"readmeSize":237},36,13017,{"closedIssues90d":239,"forks":240,"hasChangelog":233,"openIssues90d":241,"pushedAt":242,"stars":243},233,760,70,1778696315000,13231,{"hasNpmPackage":245,"license":246,"smitheryVerified":245},false,"MIT",{"hasCi":233,"hasTests":233},{"updatedAt":249},1778698625700,{"basePath":12,"githubOwner":251,"githubRepo":252,"locale":17,"slug":12,"type":223},"vectorize-io","hindsight",null,{"evaluate":255,"extract":257},{"promptVersionExtension":210,"promptVersionScoring":211,"score":214,"tags":256,"targetMarket":224,"tier":225},[217,218,219,220,221,222,223],{"commitSha":258},"HEAD",{"repoId":260},"kd7etqyey7pjacrkp7wzfh4vts86nq1z",{"_creationTime":262,"_id":260,"identity":263,"providers":264,"workflow":636},1778698349934.8708,{"githubOwner":251,"githubRepo":252,"sourceUrl":13},{"classify":265,"discover":627,"github":630},{"commitSha":258,"extensions":266},[267,284,305,312,322,330,338,346,480,487,495,507,519,530,537,574,600,609],{"basePath":268,"description":269,"displayName":252,"installMethods":270,"rationale":272,"selectedPaths":273,"source":282,"sourceLanguage":17,"type":283},"","Official Hindsight integrations for Claude Code",{"claudeCode":271},"vectorize-io/hindsight","marketplace.json at .claude-plugin/marketplace.json",[274,277,279],{"path":275,"priority":276},".claude-plugin/marketplace.json","mandatory",{"path":278,"priority":276},"README.md",{"path":280,"priority":281},"LICENSE","high","rule","marketplace",{"basePath":285,"description":286,"displayName":287,"installMethods":288,"license":246,"rationale":289,"selectedPaths":290,"source":282,"sourceLanguage":17,"type":304},"hindsight-integrations/claude-code","Automatic long-term memory for Claude Code via Hindsight. Recalls relevant memories before each prompt, retains conversation transcripts, and provides knowledge page tools.","hindsight-memory",{"claudeCode":287},"plugin manifest at hindsight-integrations/claude-code/.claude-plugin/plugin.json",[291,293,294,295,298,300,302],{"path":292,"priority":276},".claude-plugin/plugin.json",{"path":278,"priority":276},{"path":280,"priority":281},{"path":296,"priority":297},"skills/create-agent/SKILL.md","medium",{"path":299,"priority":276},".mcp.json",{"path":301,"priority":281},"hooks/hooks.json",{"path":303,"priority":276},"settings.json","plugin",{"basePath":306,"displayName":307,"installMethods":308,"rationale":309,"selectedPaths":310,"source":282,"sourceLanguage":17,"type":283},"hindsight-integrations","hindsight-local",{"claudeCode":271},"marketplace.json at hindsight-integrations/.claude-plugin/marketplace.json",[311],{"path":275,"priority":276},{"basePath":313,"description":314,"displayName":315,"installMethods":316,"rationale":317,"selectedPaths":318,"source":282,"sourceLanguage":17,"type":321},".claude/skills/code-review","Review changed code against project standards. Checks for missing tests, dead code, type safety, lint issues, and coding conventions. Run after completing any implementation work.","code-review",{"claudeCode":271},"SKILL.md frontmatter at .claude/skills/code-review/SKILL.md",[319],{"path":320,"priority":276},"SKILL.md","skill",{"basePath":323,"description":324,"displayName":325,"installMethods":326,"rationale":327,"selectedPaths":328,"source":282,"sourceLanguage":17,"type":321},"hindsight-integrations/claude-code/skills/create-agent","Create a new Hindsight-powered subagent with long-term memory. Use when the user wants a specialized agent that learns and remembers across sessions.","create-agent",{"claudeCode":271},"SKILL.md frontmatter at hindsight-integrations/claude-code/skills/create-agent/SKILL.md",[329],{"path":320,"priority":276},{"basePath":331,"description":332,"displayName":333,"installMethods":334,"rationale":335,"selectedPaths":336,"source":282,"sourceLanguage":17,"type":321},"skills/hindsight-architect","Expert memory architect. Understands your application, identifies where memory adds value, and produces an implementation plan with bank config, tag schema, and code.","hindsight-architect",{"claudeCode":271},"SKILL.md frontmatter at skills/hindsight-architect/SKILL.md",[337],{"path":320,"priority":276},{"basePath":339,"description":340,"displayName":341,"installMethods":342,"rationale":343,"selectedPaths":344,"source":282,"sourceLanguage":17,"type":321},"skills/hindsight-cloud","Store team knowledge, project conventions, and learnings from tasks. Use to remember what works and recall context before new tasks. Connects to Hindsight Cloud. (user)","hindsight-cloud",{"claudeCode":271},"SKILL.md frontmatter at skills/hindsight-cloud/SKILL.md",[345],{"path":320,"priority":276},{"basePath":347,"description":348,"displayName":349,"installMethods":350,"rationale":351,"selectedPaths":352,"source":282,"sourceLanguage":17,"type":321},"skills/hindsight-docs","Complete Hindsight documentation for AI agents. Use this to learn about Hindsight architecture, APIs, configuration, and best practices.","hindsight-docs",{"claudeCode":271},"SKILL.md frontmatter at skills/hindsight-docs/SKILL.md",[353,354,356,358,360,362,364,366,368,370,372,374,376,378,380,382,384,386,388,390,392,394,396,398,400,402,404,406,408,410,412,414,416,418,420,422,424,426,428,430,432,434,436,438,440,442,444,446,448,450,452,454,456,458,460,462,464,466,468,470,472,474,476,478],{"path":320,"priority":276},{"path":355,"priority":297},"references/best-practices.md",{"path":357,"priority":297},"references/changelog/index.md",{"path":359,"priority":297},"references/changelog/integrations/ag2.md",{"path":361,"priority":297},"references/changelog/integrations/agentcore.md",{"path":363,"priority":297},"references/changelog/integrations/ai-sdk.md",{"path":365,"priority":297},"references/changelog/integrations/autogen.md",{"path":367,"priority":297},"references/changelog/integrations/chat.md",{"path":369,"priority":297},"references/changelog/integrations/claude-code.md",{"path":371,"priority":297},"references/changelog/integrations/cloudflare-oauth-proxy.md",{"path":373,"priority":297},"references/changelog/integrations/codex.md",{"path":375,"priority":297},"references/changelog/integrations/crewai.md",{"path":377,"priority":297},"references/changelog/integrations/dify.md",{"path":379,"priority":297},"references/changelog/integrations/langgraph.md",{"path":381,"priority":297},"references/changelog/integrations/litellm.md",{"path":383,"priority":297},"references/changelog/integrations/llamaindex.md",{"path":385,"priority":297},"references/changelog/integrations/n8n.md",{"path":387,"priority":297},"references/changelog/integrations/nemoclaw.md",{"path":389,"priority":297},"references/changelog/integrations/openai-agents.md",{"path":391,"priority":297},"references/changelog/integrations/openclaw.md",{"path":393,"priority":297},"references/changelog/integrations/opencode.md",{"path":395,"priority":297},"references/changelog/integrations/paperclip.md",{"path":397,"priority":297},"references/changelog/integrations/pipecat.md",{"path":399,"priority":297},"references/changelog/integrations/pydantic-ai.md",{"path":401,"priority":297},"references/changelog/integrations/smolagents.md",{"path":403,"priority":297},"references/changelog/integrations/strands.md",{"path":405,"priority":297},"references/developer/admin-cli.md",{"path":407,"priority":297},"references/developer/api/bank-templates.md",{"path":409,"priority":297},"references/developer/api/documents.md",{"path":411,"priority":297},"references/developer/api/main-methods.md",{"path":413,"priority":297},"references/developer/api/memory-banks.md",{"path":415,"priority":297},"references/developer/api/mental-models.md",{"path":417,"priority":297},"references/developer/api/operations.md",{"path":419,"priority":297},"references/developer/api/quickstart.md",{"path":421,"priority":297},"references/developer/api/recall.md",{"path":423,"priority":297},"references/developer/api/reflect.md",{"path":425,"priority":297},"references/developer/api/retain.md",{"path":427,"priority":297},"references/developer/api/webhooks.md",{"path":429,"priority":297},"references/developer/configuration.md",{"path":431,"priority":297},"references/developer/development.md",{"path":433,"priority":297},"references/developer/extensions.md",{"path":435,"priority":297},"references/developer/index.md",{"path":437,"priority":297},"references/developer/installation.md",{"path":439,"priority":297},"references/developer/mcp-server.md",{"path":441,"priority":297},"references/developer/models.md",{"path":443,"priority":297},"references/developer/monitoring.md",{"path":445,"priority":297},"references/developer/multilingual.md",{"path":447,"priority":297},"references/developer/observations.md",{"path":449,"priority":297},"references/developer/performance.md",{"path":451,"priority":297},"references/developer/rag-vs-hindsight.md",{"path":453,"priority":297},"references/developer/reflect.md",{"path":455,"priority":297},"references/developer/retain.md",{"path":457,"priority":297},"references/developer/retrieval.md",{"path":459,"priority":297},"references/developer/services.md",{"path":461,"priority":297},"references/developer/storage.md",{"path":463,"priority":297},"references/faq.md",{"path":465,"priority":297},"references/openapi.json",{"path":467,"priority":297},"references/sdks/cli.md",{"path":469,"priority":297},"references/sdks/embed.md",{"path":471,"priority":297},"references/sdks/go.md",{"path":473,"priority":297},"references/sdks/hindsight-all-npm.md",{"path":475,"priority":297},"references/sdks/hindsight-all.md",{"path":477,"priority":297},"references/sdks/nodejs.md",{"path":479,"priority":297},"references/sdks/python.md",{"basePath":481,"description":482,"displayName":307,"installMethods":483,"rationale":484,"selectedPaths":485,"source":282,"sourceLanguage":17,"type":321},"skills/hindsight-local","Store user preferences, learnings from tasks, and procedure outcomes. Use to remember what works and recall context before new tasks. (user)",{"claudeCode":271},"SKILL.md frontmatter at skills/hindsight-local/SKILL.md",[486],{"path":320,"priority":276},{"basePath":488,"description":489,"displayName":490,"installMethods":491,"rationale":492,"selectedPaths":493,"source":282,"sourceLanguage":17,"type":321},"skills/hindsight-self-hosted","Store team knowledge, project conventions, and learnings from tasks. Use to remember what works and recall context before new tasks. Connects to a self-hosted Hindsight server. (user)","hindsight-self-hosted",{"claudeCode":271},"SKILL.md frontmatter at skills/hindsight-self-hosted/SKILL.md",[494],{"path":320,"priority":276},{"basePath":496,"description":497,"displayName":498,"installMethods":499,"license":500,"rationale":501,"selectedPaths":502,"source":282,"sourceLanguage":17,"type":223},"hindsight-control-plane","Control plane for Hindsight - Semantic memory system","@vectorize-io/hindsight-control-plane",{"npm":498},"ISC","cli ecosystem detected at hindsight-control-plane",[503,505],{"path":504,"priority":276},"package.json",{"path":506,"priority":297},"bin/cli.js",{"basePath":508,"description":509,"displayName":510,"installMethods":511,"license":246,"rationale":512,"selectedPaths":513,"source":282,"sourceLanguage":17,"type":223},"hindsight-integrations/nemoclaw","Setup CLI for hindsight-openclaw on NemoClaw sandboxes — installs the plugin, configures external API mode, and applies the OpenShell network policy","@vectorize-io/hindsight-nemoclaw",{"npm":510},"cli ecosystem detected at hindsight-integrations/nemoclaw",[514,515,516],{"path":504,"priority":276},{"path":278,"priority":276},{"path":517,"priority":518},"src/cli.ts","low",{"basePath":520,"description":521,"displayName":522,"installMethods":523,"license":246,"rationale":524,"selectedPaths":525,"source":282,"sourceLanguage":17,"type":223},"hindsight-integrations/openclaw","Hindsight memory plugin for OpenClaw - biomimetic long-term memory with fact extraction","@vectorize-io/hindsight-openclaw",{"npm":522},"cli ecosystem detected at hindsight-integrations/openclaw",[526,527,528],{"path":504,"priority":276},{"path":278,"priority":276},{"path":529,"priority":518},"src/index.ts",{"basePath":12,"installMethods":531,"rationale":532,"selectedPaths":533,"source":282,"sourceLanguage":17,"type":223},{"pypi":12},"cli ecosystem detected at hindsight-api",[534,536],{"path":535,"priority":276},"pyproject.toml",{"path":278,"priority":276},{"basePath":538,"installMethods":539,"rationale":540,"selectedPaths":541,"source":282,"sourceLanguage":17,"type":223},"hindsight-cli",{"cargo":538},"cli ecosystem detected at hindsight-cli",[542,544,546,548,550,552,554,556,558,560,562,564,566,568,570,572],{"path":543,"priority":276},"Cargo.toml",{"path":545,"priority":297},"src/main.rs",{"path":547,"priority":518},"src/commands/audit.rs",{"path":549,"priority":518},"src/commands/bank.rs",{"path":551,"priority":518},"src/commands/chunk.rs",{"path":553,"priority":518},"src/commands/directive.rs",{"path":555,"priority":518},"src/commands/document.rs",{"path":557,"priority":518},"src/commands/entity.rs",{"path":559,"priority":518},"src/commands/explore.rs",{"path":561,"priority":518},"src/commands/health.rs",{"path":563,"priority":518},"src/commands/memory.rs",{"path":565,"priority":518},"src/commands/mental_model.rs",{"path":567,"priority":518},"src/commands/mod.rs",{"path":569,"priority":518},"src/commands/operation.rs",{"path":571,"priority":518},"src/commands/tag.rs",{"path":573,"priority":518},"src/commands/webhook.rs",{"basePath":575,"installMethods":576,"rationale":577,"selectedPaths":578,"source":282,"sourceLanguage":599,"type":223},"hindsight-dev",{"pypi":575},"cli ecosystem detected at hindsight-dev",[579,580,581,583,585,587,589,591,593,595,597],{"path":535,"priority":276},{"path":278,"priority":276},{"path":582,"priority":297},"hindsight_dev/generate_openapi.py",{"path":584,"priority":297},"hindsight_dev/generate_bank_template_schema.py",{"path":586,"priority":297},"hindsight_dev/generate_changelog.py",{"path":588,"priority":297},"hindsight_dev/sync_cookbook.py",{"path":590,"priority":297},"hindsight_dev/generate_llms_full.py",{"path":592,"priority":297},"hindsight_dev/check_openapi_compatibility.py",{"path":594,"priority":297},"hindsight_dev/cli_coverage_check.py",{"path":596,"priority":297},"hindsight_dev/client_coverage_check.py",{"path":598,"priority":297},"benchmarks/perf/system_perf.py","sv",{"basePath":601,"installMethods":602,"rationale":603,"selectedPaths":604,"source":282,"sourceLanguage":17,"type":223},"hindsight-embed",{"pypi":601},"cli ecosystem detected at hindsight-embed",[605,606,607],{"path":535,"priority":276},{"path":278,"priority":276},{"path":608,"priority":297},"hindsight_embed/cli.py",{"basePath":610,"displayName":610,"installMethods":611,"rationale":612,"selectedPaths":613,"source":282,"sourceLanguage":17,"type":626},"hindsight-api-slim",{"pypi":610},"pyproject.toml with mcp/fastmcp dependency + scripts at hindsight-api-slim/pyproject.toml",[614,615,616,618,620,622,624],{"path":535,"priority":276},{"path":278,"priority":276},{"path":617,"priority":297},"hindsight_api/main.py",{"path":619,"priority":297},"hindsight_api/worker/main.py",{"path":621,"priority":297},"hindsight_api/mcp_local.py",{"path":623,"priority":297},"hindsight_api/admin/cli.py",{"path":625,"priority":518},"hindsight_api/server.py","mcp",{"sources":628},[629],"manual",{"closedIssues90d":239,"description":10,"forks":240,"homepage":631,"license":246,"openIssues90d":241,"pushedAt":242,"readmeSize":237,"stars":243,"topics":632},"https://hindsight.vectorize.io/",[633,218,634,635],"agentic-ai","agents","ai-memory",{"classifiedAt":637,"discoverAt":638,"extractAt":639,"githubAt":639,"updatedAt":637},1778698371538,1778698349934,1778698367862,[219,217,223,221,220,218,222],{"evaluatedAt":249,"extractAt":642,"updatedAt":249},1778698371743,[],[645,677,704],{"_creationTime":646,"_id":647,"community":648,"display":649,"identity":655,"providers":658,"relations":669,"tags":672,"workflow":673},1778692566881.227,"k170x6w0548s68g79ejga9a6a186mvtw",{"reviewCount":8},{"description":650,"installMethods":651,"name":653,"sourceUrl":654},"用于 MCP 安全的开源 AI 代理防火墙：代理出口控制、DLP、SSRF 和提示注入防护。",{"go":652},"github.com/luckyPipewrench/pipelock","pipelock","https://github.com/luckyPipewrench/pipelock",{"basePath":268,"githubOwner":656,"githubRepo":653,"locale":657,"slug":653,"type":223},"luckyPipewrench","zh-CN",{"evaluate":659,"extract":668},{"promptVersionExtension":210,"promptVersionScoring":211,"score":660,"tags":661,"targetMarket":224,"tier":225},100,[662,663,219,223,626,664,665,666,667],"security","firewall","dlp","ssrf","injection-defense","auditing",{"commitSha":258},{"repoId":670,"translatedFrom":671},"kd728h5jcjjje6rbqwymfj9zph86m4jp","k17bc73bymk3wvfgy6mjgpjcps86ncc6",[219,667,223,664,663,666,626,662,665],{"evaluatedAt":674,"extractAt":675,"updatedAt":676},1778692494031,1778692456956,1778692566881,{"_creationTime":678,"_id":679,"community":680,"display":681,"identity":686,"providers":690,"relations":698,"tags":700,"workflow":701},1778683644393.5793,"k171esc5b8pbh1r9s90d526z3586njrn",{"reviewCount":8},{"description":682,"installMethods":683,"name":684,"sourceUrl":685},"Local skill manager and installer for agent-playbook across Claude Code, Codex, and Gemini.",{"npm":684},"@codeharbor/agent-playbook","https://github.com/charon-fan/agent-playbook",{"basePath":687,"githubOwner":688,"githubRepo":689,"locale":17,"slug":689,"type":223},"packages/agent-playbook","charon-fan","agent-playbook",{"evaluate":691,"extract":697},{"promptVersionExtension":210,"promptVersionScoring":211,"score":660,"tags":692,"targetMarket":224,"tier":225},[223,219,693,694,695,696],"skills","management","automation","configuration",{"commitSha":258,"license":246},{"repoId":699},"kd7cf5d43dzccs0fw9c2rp131n86mrv4",[219,695,223,696,694,693],{"evaluatedAt":702,"extractAt":703,"updatedAt":702},1778684270439,1778683644393,{"_creationTime":705,"_id":706,"community":707,"display":708,"identity":714,"providers":717,"relations":725,"tags":728,"workflow":729},1778697166802.8904,"k17brwp2rz821bkz2zeazsv5td86mdep",{"reviewCount":8},{"description":709,"installMethods":710,"name":712,"sourceUrl":713},"AI约束引擎 — 强制执行 CLAUDE.md、.cursorrules、AGENTS.md 规则，如同法律一般。拥有 51 个 MCP 工具，991 个测试。官方 MCP 注册表。 npx speclock protect",{"pypi":711},"speclock_ros2","speclock-ros2","https://github.com/sgroy10/speclock",{"basePath":712,"githubOwner":715,"githubRepo":716,"locale":657,"slug":712,"type":223},"sgroy10","speclock",{"evaluate":718,"extract":724},{"promptVersionExtension":210,"promptVersionScoring":211,"score":660,"tags":719,"targetMarket":224,"tier":225},[720,721,722,223,662,626,723],"ai-constraints","code-enforcement","developer-tool","robotics",{"commitSha":258},{"repoId":726,"translatedFrom":727},"kd73m85wtkj05xmfff4vq695hn86mgg4","k177h2k24qprbp2pjjs5tsb9r586nywg",[720,223,721,722,626,723,662],{"evaluatedAt":730,"extractAt":731,"updatedAt":732},1778697123911,1778697107735,1778697166803]