{"passport":{"unfragile":{"@version":"1.0","version":"2026-05","artifact":{"id":"github-jnmetacode--agency-agents-zh","slug":"jnmetacode--agency-agents-zh","name":"agency-agents-zh","type":"agent","url":"https://github.com/jnMetaCode/agency-orchestrator","page_url":"https://unfragile.ai/jnmetacode--agency-agents-zh","categories":["ai-agents","code-editors","app-builders"],"tags":["agency-orchestrator","agent-definitions","ai-agents","ai-roles","chinese","claude","claude-code","copilot-agent","cursor-rules","deepseek","hermes-agent","llm","multi-agent","no-code","prompt-engineering","qwen","system-prompt","workbuddy","workflow"],"pricing":{"model":"open_source","free":true,"starting_price":null},"status":"active","verified":false},"capabilities":[{"id":"github-jnmetacode--agency-agents-zh__cap_0","uri":"capability://automation.workflow.multi.tool.agent.deployment.pipeline.with.format.auto.conversion","name":"multi-tool agent deployment pipeline with format auto-conversion","description":"Converts unified Markdown-based agent definitions (with YAML frontmatter) into tool-specific formats via a two-stage bash/PowerShell pipeline (convert.sh → install.sh). The convert stage parses raw agent files and transforms them into 14+ target formats (.mdc for Cursor, .json for Kiro, aggregated files for Aider/Windsurf, rules for Claude Code/Copilot). The install stage auto-detects local tool installations and deploys converted agents to the correct configuration directories (~/.claude/agents/, .cursor/rules/, etc.), eliminating manual file placement.","intents":["Deploy the same agent definition across multiple AI development tools without manual reformatting","Automatically detect which tools are installed locally and populate their agent directories","Maintain a single source of truth for agent definitions while supporting heterogeneous tool ecosystems","Reduce friction when switching between Cursor, Claude Code, GitHub Copilot, Windsurf, and other IDE integrations"],"best_for":["Teams using multiple AI-assisted development tools (Cursor + Claude Code + Copilot)","Organizations standardizing on agent definitions across departments","Developers who want to version-control agents in git without tool-specific lock-in"],"limitations":["Conversion fidelity depends on target tool's schema — some advanced features may not translate (e.g., Claude Code's native memory features may not map to Cursor rules)","Bash pipeline requires Unix-like environment; PowerShell variant adds Windows support but requires separate maintenance","No real-time sync — agents must be re-deployed after updates; no hot-reload capability","Tool detection is filesystem-based (checking ~/.cursor/, ~/.claude/, etc.) and may fail with non-standard installations"],"requires":["Bash 4.0+ or PowerShell 5.0+","At least one supported AI tool installed (Cursor, Claude Code, GitHub Copilot, Windsurf, Aider, etc.)","Read/write permissions to tool configuration directories","Git (optional, for version control of agent definitions)"],"input_types":["Markdown files with YAML frontmatter (agent definitions)","Directory structure following department taxonomy (engineering/, design/, marketing/, etc.)"],"output_types":[".mdc files (Cursor rules format)",".json files (Kiro/tool-agnostic format)","Aggregated single-file formats (Aider, Windsurf)","System prompt strings (Claude Code, GitHub Copilot)","Bash/shell scripts (direct integration)"],"categories":["automation-workflow","tool-use-integration"],"confidence":0.5,"matches":0,"success_rate":0},{"id":"github-jnmetacode--agency-agents-zh__cap_1","uri":"capability://memory.knowledge.structured.agent.library.with.18.department.taxonomy.and.chinese.market.specialization","name":"structured agent library with 18-department taxonomy and chinese market specialization","description":"Organizes 211 pre-built AI agent personas across 18 professional departments (Engineering, Design, Marketing, Sales, Support, Testing, Finance, Legal, HR, Academic, Game Development, Supply Chain, etc.) with 46 agents specifically designed for Chinese platforms (Xiaohongshu, Douyin, WeChat, Feishu, DingTalk). Each agent is defined as a Markdown file containing YAML metadata (name, department, tools, version) and a structured body (identity/mission/rules/deliverables). The library is indexed and linted via CI/CD to ensure consistency and completeness.","intents":["Quickly find and deploy a pre-configured AI agent for a specific professional role without writing prompts from scratch","Access domain-specific agents optimized for Chinese market platforms and business contexts","Understand what cognitive frameworks and workflows each agent applies to its domain","Contribute new agents to the library following standardized templates and validation rules"],"best_for":["Chinese-speaking teams and organizations building AI-assisted workflows","Enterprises needing role-specific agents (e.g., 'Xiaohongshu Content Strategist', 'DingTalk Workflow Optimizer')","Developers building multi-agent systems who want pre-vetted, production-ready personas","Non-technical users who want to deploy agents without understanding prompt engineering"],"limitations":["Library is static — agents are not dynamically updated based on user feedback or performance metrics","Chinese market agents may not generalize well to non-Chinese contexts (e.g., Xiaohongshu strategies don't apply to TikTok)","No built-in mechanism to customize agents per-organization (e.g., company-specific tone, policies) — requires forking or manual editing","Agents are persona-based, not task-based — may require chaining multiple agents for complex workflows"],"requires":["Access to the GitHub repository (public, no authentication required)","Basic understanding of Markdown and YAML syntax (for reading/editing agent definitions)","One of the 14+ supported AI tools (Cursor, Claude Code, Copilot, Windsurf, etc.)"],"input_types":["Agent name or department category (text search)","Use case description (semantic matching against agent missions)"],"output_types":["Agent definition (Markdown + YAML)","Deployed agent in target tool (after conversion/installation)","Agent metadata (JSON catalog)"],"categories":["memory-knowledge","planning-reasoning"],"confidence":0.5,"matches":0,"success_rate":0},{"id":"github-jnmetacode--agency-agents-zh__cap_10","uri":"capability://tool.use.integration.openclaw.workspace.integration.for.unified.agent.deployment","name":"openclaw workspace integration for unified agent deployment","description":"Integrates with OpenClaw (a workspace management tool) to enable unified deployment and management of agents across multiple tools and projects. OpenClaw provides a centralized interface for selecting, configuring, and deploying agents to local development environments. The integration leverages the conversion pipeline to automatically deploy agents to the correct tool-specific formats and directories. This reduces friction for teams that use multiple tools and want a single point of control for agent deployment.","intents":["Deploy agents to multiple tools from a single OpenClaw interface","Manage agent configurations and versions centrally","Enable team members to quickly set up their development environment with the correct agents","Track which agents are deployed to which tools and projects"],"best_for":["Teams using OpenClaw for workspace management","Organizations with multiple developers using different tools (Cursor, Claude Code, Copilot, etc.)","Teams that want centralized control over agent deployment","Enterprises standardizing on agent definitions across projects"],"limitations":["OpenClaw integration is optional and requires OpenClaw to be installed and configured","OpenClaw may not support all tools in the agency-agents-zh ecosystem","Deployment is one-way (OpenClaw → tools); changes in tools are not synced back to OpenClaw","OpenClaw configuration is tool-specific; may require manual setup for each tool","No built-in version control or rollback — agent updates may break existing workflows"],"requires":["OpenClaw installed and configured","Integration configuration (which agents to deploy, target tools)","Conversion pipeline (convert.sh/install.sh) to generate tool-specific formats"],"input_types":["Agent selection (which agents to deploy)","Tool selection (which tools to deploy to)","Workspace configuration (project directories, tool paths)"],"output_types":["Deployed agents in target tools","Deployment status and logs","Workspace configuration file (for version control)"],"categories":["tool-use-integration","automation-workflow"],"confidence":0.5,"matches":0,"success_rate":0},{"id":"github-jnmetacode--agency-agents-zh__cap_11","uri":"capability://planning.reasoning.scenario.based.runbook.templates.for.common.workflows","name":"scenario-based runbook templates for common workflows","description":"Provides pre-defined scenario runbooks (e.g., 'xiaohongshu-launch', 'product-development', 'infrastructure-deployment') that orchestrate multiple agents through a complete workflow. Each runbook specifies the sequence of agents, handoff protocols, validation checkpoints, and expected outputs. Runbooks are defined in Markdown and can be executed via the NEXUS orchestration framework or manually. This enables teams to standardize on repeatable workflows without building orchestration logic from scratch.","intents":["Execute a complete workflow (e.g., product launch) by following a pre-defined runbook","Ensure consistent execution of complex workflows across team members","Reduce time spent on workflow planning and agent coordination","Capture institutional knowledge about how to execute common tasks"],"best_for":["Teams with repeatable workflows (product launches, content creation, infrastructure deployment)","Organizations that want to standardize on best practices","Developers building multi-agent systems who want templates to start from","Non-technical users who want to execute complex workflows without understanding agent orchestration"],"limitations":["Runbooks are pre-defined and static — cannot adapt to unexpected situations or deviations","Runbooks assume a specific sequence of agents; parallel or conditional workflows may not fit the model","Runbooks are text-based; no visual workflow editor or drag-and-drop interface","Runbooks do not include error handling or recovery logic — failures may require manual intervention","Runbooks are organization-agnostic; may require customization for specific company processes or policies"],"requires":["Understanding of the runbook format (Markdown + YAML)","Agents defined in the library (for the agents referenced in the runbook)","NEXUS orchestration framework (for programmatic execution) or manual execution capability"],"input_types":["Runbook name (e.g., 'xiaohongshu-launch')","Initial context (e.g., product description, target audience)","Customization parameters (e.g., timeline, budget, team size)"],"output_types":["Workflow execution logs","Intermediate deliverables from each agent","Final deliverable (e.g., launch plan, content calendar, deployment checklist)"],"categories":["planning-reasoning","automation-workflow"],"confidence":0.5,"matches":0,"success_rate":0},{"id":"github-jnmetacode--agency-agents-zh__cap_12","uri":"capability://automation.workflow.agent.contribution.framework.with.standardized.templates","name":"agent contribution framework with standardized templates","description":"Provides standardized templates and contribution guidelines for adding new agents to the library. Contributors create a Markdown file with YAML frontmatter (metadata) and a structured body (identity, mission, rules, deliverables) following the template. The contribution process includes validation via the linter, peer review, and integration into the appropriate department. The framework ensures consistency across all 211 agents and makes it easy for community members to contribute without understanding the entire codebase.","intents":["Contribute a new agent to the library following a standardized template","Ensure new agents meet quality standards and are consistent with existing agents","Understand what information is required to define an agent","Get feedback on agent definitions before they are merged"],"best_for":["Community members who want to contribute new agents to the library","Organizations that want to extend the library with domain-specific agents","Teams that want to standardize on agent definitions across projects","Maintainers who want to ensure quality and consistency in contributions"],"limitations":["Contribution process requires GitHub account and familiarity with git/GitHub","Peer review may be slow depending on maintainer availability","Templates are prescriptive — may not accommodate agents with non-standard structures","No built-in mechanism to deprecate or remove agents — library may accumulate outdated agents","Contribution guidelines are in English and Chinese; non-English speakers may have difficulty"],"requires":["GitHub account","Git installed locally","Understanding of Markdown and YAML syntax","Familiarity with the agent definition template","Understanding of the department taxonomy"],"input_types":["Agent definition (Markdown + YAML, following the template)","Agent metadata (name, department, tools, version)","Agent identity, mission, rules, and deliverables (text)"],"output_types":["Pull request with the new agent definition","Linter validation report","Peer review feedback","Merged agent in the library"],"categories":["automation-workflow","planning-reasoning"],"confidence":0.5,"matches":0,"success_rate":0},{"id":"github-jnmetacode--agency-agents-zh__cap_2","uri":"capability://planning.reasoning.nexus.multi.agent.orchestration.with.phase.based.lifecycle.and.handoff.protocols","name":"nexus multi-agent orchestration with phase-based lifecycle and handoff protocols","description":"Implements a 7-phase orchestration framework (Phases 0–6) for coordinating multiple agents on complex tasks. Each phase defines entry conditions, agent responsibilities, handoff protocols, and validation checkpoints. Agents communicate via standardized handoff templates that specify context, constraints, and expected outputs. The framework includes scenario runbooks (pre-defined workflows for common patterns like 'product launch', 'content creation', 'infrastructure deployment') and agent coordination templates that define who hands off to whom and under what conditions. Implemented via the agency-orchestrator NPM package for programmatic control.","intents":["Orchestrate multiple specialized agents (e.g., 'Product Manager' → 'Engineering Lead' → 'QA Tester') on a single complex project","Define repeatable workflows (runbooks) that automatically route tasks to the right agent based on phase and context","Ensure agents hand off work with sufficient context and validation to prevent rework","Track progress across a multi-agent workflow and identify bottlenecks or failures"],"best_for":["Teams building multi-agent AI systems for product development, content creation, or infrastructure management","Organizations with repeatable workflows (e.g., 'launch a new product', 'create a marketing campaign') that benefit from standardization","Developers who want to avoid ad-hoc agent chaining and prefer declarative, auditable orchestration"],"limitations":["Phase-based model assumes sequential or loosely-parallel workflows; highly concurrent or dynamic task graphs may not fit the model","Handoff protocols are text-based (no native state machine or formal verification) — relies on agent compliance and manual validation","No built-in feedback loops or learning — agents don't adapt based on previous handoff failures","Runbooks are pre-defined; dynamic workflow generation based on user input is not supported","Requires NPM package (agency-orchestrator) for programmatic use; no native REST API or webhook support"],"requires":["Node.js 14+ (for agency-orchestrator NPM package)","Understanding of the 7-phase lifecycle model and handoff protocol conventions","At least 2 agents defined in the library (for meaningful orchestration)","Optional: MCP (Model Context Protocol) for memory integration"],"input_types":["Scenario name (e.g., 'xiaohongshu-launch') to trigger a runbook","Task description and initial context (text)","Agent definitions and phase configurations (YAML/JSON)"],"output_types":["Phase execution logs (text/JSON)","Handoff records with context and validation results","Final deliverable from the last agent in the chain","Workflow execution trace (for debugging)"],"categories":["planning-reasoning","automation-workflow"],"confidence":0.5,"matches":0,"success_rate":0},{"id":"github-jnmetacode--agency-agents-zh__cap_3","uri":"capability://tool.use.integration.rules.based.agent.integration.for.ide.native.tools.cursor.trae.opencode","name":"rules-based agent integration for ide-native tools (cursor, trae, opencode)","description":"Converts agent definitions into .mdc (Markdown with Code) rule files for Cursor, Trae, and OpenCode, which use a rules-based system to inject agent personas into the IDE's code completion and suggestion engine. The conversion process extracts the agent's identity, mission, and rules from the Markdown definition and formats them as a .cursor/rules/ file that the IDE loads at startup. Rules are applied contextually — the IDE evaluates them against the current file, selection, and command to determine which agent persona should influence suggestions. This enables IDE-native agent switching without leaving the editor.","intents":["Switch between specialized agent personas (e.g., 'Senior Developer', 'Mobile App Builder', 'Security Auditor') within the IDE without manual prompt engineering","Have IDE suggestions and code completions reflect the agent's expertise and coding standards","Maintain consistent agent behavior across multiple IDE sessions and projects","Avoid context switching between the IDE and external agent tools"],"best_for":["Cursor, Trae, or OpenCode users who want to leverage pre-built agent personas","Teams standardizing on a specific IDE and wanting consistent agent behavior across developers","Developers who prefer IDE-native workflows over external agent tools"],"limitations":["Rules-based integration is IDE-specific; agents defined for Cursor may not work in Trae without re-conversion","IDE rule evaluation is opaque — no visibility into why a particular suggestion was generated or which rule triggered it","Rules are static; no dynamic rule generation based on project context or user feedback","Limited to IDE-native capabilities — cannot invoke external APIs or tools beyond the IDE's built-in functions","Rule file size is limited by IDE constraints; very complex agents may not fit in a single .mdc file"],"requires":["Cursor 0.30+, Trae, or OpenCode installed","Write permissions to .cursor/rules/ or equivalent directory","Agent definition in the library (Markdown + YAML format)"],"input_types":["Agent definition (Markdown + YAML)","IDE context (current file, selection, command)"],"output_types":[".mdc rule file (Cursor/Trae/OpenCode format)","IDE suggestions and code completions influenced by the agent persona"],"categories":["tool-use-integration","code-generation-editing"],"confidence":0.5,"matches":0,"success_rate":0},{"id":"github-jnmetacode--agency-agents-zh__cap_4","uri":"capability://tool.use.integration.direct.copy.agent.integration.for.claude.code.and.github.copilot","name":"direct-copy agent integration for claude code and github copilot","description":"Provides agent definitions formatted as copy-paste system prompts for Claude Code and GitHub Copilot, which do not support external rule files or configuration. The conversion process extracts the agent's identity, mission, and rules from the Markdown definition and formats them as a plain-text system prompt that can be pasted directly into Claude Code's system prompt field or GitHub Copilot's settings. This is the simplest integration path but requires manual copy-paste and does not support dynamic agent switching within a session.","intents":["Use pre-built agent personas in Claude Code or GitHub Copilot without writing custom system prompts","Quickly switch between agents by copy-pasting different system prompts","Maintain consistency across team members using the same agent definitions","Version-control agent definitions in git and deploy them via copy-paste"],"best_for":["Claude Code and GitHub Copilot users who want pre-built agent personas","Teams that prefer simplicity over automation (copy-paste is more transparent than automated deployment)","Developers who want to inspect and customize agent prompts before use"],"limitations":["Requires manual copy-paste for each agent switch — no automation or IDE integration","No dynamic agent switching within a session — must restart the tool to change agents","System prompt length is limited by the tool's input constraints; very complex agents may exceed limits","No feedback loop — cannot track which agent was used or measure effectiveness","Prone to human error (copy-paste mistakes, version mismatches)"],"requires":["Claude Code or GitHub Copilot installed","Access to the agent definition (Markdown or converted system prompt)","Manual copy-paste capability"],"input_types":["Agent definition (Markdown + YAML or pre-converted system prompt text)"],"output_types":["Plain-text system prompt (copy-paste format)","Agent behavior in Claude Code or GitHub Copilot"],"categories":["tool-use-integration","text-generation-language"],"confidence":0.5,"matches":0,"success_rate":0},{"id":"github-jnmetacode--agency-agents-zh__cap_5","uri":"capability://tool.use.integration.single.file.agent.aggregation.for.aider.and.windsurf","name":"single-file agent aggregation for aider and windsurf","description":"Converts multiple agent definitions into a single aggregated file (e.g., agents.md or agents.json) that Aider and Windsurf can load as a unified agent catalog. The aggregation process concatenates agent definitions while preserving metadata and ensuring no conflicts. Aider and Windsurf then parse this file to populate their agent selection UI, allowing users to switch between agents without restarting the tool. This approach balances automation (single file deployment) with simplicity (no complex configuration).","intents":["Deploy all 211 agents to Aider or Windsurf in a single file","Enable dynamic agent switching within Aider/Windsurf without tool restart","Maintain a single source of truth for agent definitions across multiple tools","Reduce deployment complexity by aggregating agents into a standard format"],"best_for":["Aider and Windsurf users who want access to the full agent library","Teams using Aider/Windsurf as their primary AI-assisted development tool","Developers who want dynamic agent switching without manual copy-paste"],"limitations":["Aggregated file size grows with the number of agents — may impact tool performance if the file exceeds tool limits","Tool support for agent catalogs is tool-specific; Aider and Windsurf may have different parsing requirements","No per-agent versioning — all agents in the aggregated file must be updated together","Aggregation process is one-way; changes to individual agent files require re-aggregation and re-deployment","Tool may cache the aggregated file; updates may not be reflected immediately"],"requires":["Aider or Windsurf installed","Write permissions to the tool's agent configuration directory","Agent definitions in the library (Markdown + YAML format)"],"input_types":["Multiple agent definitions (Markdown + YAML)","Aggregation configuration (which agents to include, output format)"],"output_types":["Single aggregated file (agents.md, agents.json, or tool-specific format)","Agent catalog in Aider/Windsurf UI"],"categories":["tool-use-integration","data-processing-analysis"],"confidence":0.5,"matches":0,"success_rate":0},{"id":"github-jnmetacode--agency-agents-zh__cap_6","uri":"capability://tool.use.integration.skill.based.agent.integration.for.antigravity.and.gemini.cli","name":"skill-based agent integration for antigravity and gemini cli","description":"Converts agent definitions into skill definitions for Antigravity and Gemini CLI, which use a skill-based system to extend their capabilities. Each agent is mapped to a skill (e.g., 'Senior Developer' → 'advanced-coding-skill', 'Content Strategist' → 'content-creation-skill') with associated metadata (description, parameters, examples). The conversion process extracts the agent's mission and rules and formats them as skill definitions that Antigravity/Gemini CLI can invoke. This enables agents to be used as reusable skills that can be chained or composed.","intents":["Use agents as reusable skills in Antigravity or Gemini CLI workflows","Compose multiple agent skills into complex workflows","Enable skill discovery and documentation within Antigravity/Gemini CLI","Leverage agent expertise as a building block for larger automation systems"],"best_for":["Antigravity and Gemini CLI users building complex automation workflows","Teams that want to compose agents into larger systems","Developers who prefer skill-based composition over direct agent invocation"],"limitations":["Skill-based integration is tool-specific; skills defined for Antigravity may not work in Gemini CLI without re-conversion","Skill composition is limited by tool capabilities — complex workflows may require custom skill implementations","No built-in skill versioning or dependency management — skill updates may break dependent workflows","Skill parameters must be explicitly defined; agents with dynamic or context-dependent parameters may not map well to skills","Limited visibility into skill execution — no built-in logging or debugging for skill chains"],"requires":["Antigravity or Gemini CLI installed","Understanding of the tool's skill definition format","Agent definitions in the library (Markdown + YAML format)"],"input_types":["Agent definition (Markdown + YAML)","Skill composition configuration (which agents to expose as skills, parameters)"],"output_types":["Skill definition (tool-specific format)","Skill invocation results (text, code, or structured data)"],"categories":["tool-use-integration","planning-reasoning"],"confidence":0.5,"matches":0,"success_rate":0},{"id":"github-jnmetacode--agency-agents-zh__cap_7","uri":"capability://safety.moderation.agent.linting.and.validation.with.ci.cd.integration","name":"agent linting and validation with ci/cd integration","description":"Implements a bash-based linter (lint-agents.sh) that validates agent definitions against a schema (YAML frontmatter structure, required fields, formatting consistency). The linter checks for missing metadata (name, department, version), malformed YAML, incomplete rules sections, and inconsistent formatting. It outputs validation errors and warnings that can be integrated into GitHub Actions workflows, preventing invalid agents from being merged into the repository. This ensures library quality and consistency without manual review.","intents":["Automatically validate agent definitions before they are merged into the library","Catch common errors (missing fields, malformed YAML) early in the contribution process","Enforce consistent formatting and structure across all 211 agents","Provide feedback to contributors on what needs to be fixed"],"best_for":["Maintainers of the agency-agents-zh library who want to enforce quality standards","Teams contributing new agents and wanting automated validation feedback","CI/CD pipelines that need to validate agent definitions before deployment"],"limitations":["Linter is schema-based only — cannot validate semantic correctness (e.g., whether an agent's mission is realistic or well-written)","Linter is bash-based; requires Unix-like environment (no native Windows support, though PowerShell variant exists)","Linter output is text-based; no structured error reporting (e.g., JSON) for programmatic consumption","Linter does not check for duplicate agents or conflicting definitions across departments","Linter cannot validate agent behavior or effectiveness — only structure"],"requires":["Bash 4.0+","Agent definitions in Markdown + YAML format","Optional: GitHub Actions for CI/CD integration"],"input_types":["Agent definition files (Markdown + YAML)","Directory structure (agents organized by department)"],"output_types":["Validation report (text, with error/warning counts)","Exit code (0 for success, non-zero for failure)","GitHub Actions workflow status (pass/fail)"],"categories":["safety-moderation","automation-workflow"],"confidence":0.5,"matches":0,"success_rate":0},{"id":"github-jnmetacode--agency-agents-zh__cap_8","uri":"capability://text.generation.language.chinese.market.platform.specific.agents.with.localized.workflows","name":"chinese market platform-specific agents with localized workflows","description":"Provides 46 agents specifically designed for Chinese platforms (Xiaohongshu, Douyin, WeChat, Feishu, DingTalk) with localized workflows, content strategies, and platform-specific best practices. Each agent understands platform-specific features (e.g., Xiaohongshu's hashtag system, Douyin's algorithm, WeChat's group dynamics) and provides guidance tailored to that platform. Agents include 'Xiaohongshu Content Strategist', 'Douyin Trend Analyzer', 'WeChat Community Manager', 'Feishu Workflow Optimizer', and 'DingTalk HR Specialist'. These agents are integrated into the same deployment pipeline as generic agents, enabling seamless use across tools.","intents":["Create content optimized for Xiaohongshu, Douyin, WeChat, or other Chinese platforms without manual research","Understand platform-specific algorithms, trends, and best practices from an AI expert","Manage workflows and teams using Feishu or DingTalk with AI assistance","Localize global marketing or product strategies to Chinese market dynamics"],"best_for":["Teams and creators targeting Chinese audiences on Xiaohongshu, Douyin, WeChat, etc.","Enterprises using Feishu or DingTalk for internal workflows and wanting AI assistance","Global companies localizing products or marketing to the Chinese market","Content creators and marketers who want to leverage platform-specific insights"],"limitations":["Agents are based on historical platform data and may not reflect real-time algorithm changes","Platform-specific strategies may become outdated as platforms evolve (e.g., Douyin's algorithm changes)","Agents cannot directly interact with platforms (e.g., cannot post to Xiaohongshu) — they provide guidance only","Localization quality depends on the agent's training data; some nuances may be missed","Agents are in Chinese; non-Chinese speakers may have difficulty using them effectively"],"requires":["Account on the target platform (Xiaohongshu, Douyin, WeChat, Feishu, DingTalk) or understanding of its features","One of the 14+ supported AI tools (Cursor, Claude Code, Copilot, etc.)","Basic understanding of Chinese market dynamics (helpful but not required)"],"input_types":["Content idea, product description, or workflow requirement (text)","Platform name (to select the appropriate agent)","Target audience or campaign goals (text)"],"output_types":["Platform-optimized content strategy or guidance (text)","Specific recommendations (hashtags, posting times, content format)","Workflow optimization suggestions (for Feishu/DingTalk agents)"],"categories":["text-generation-language","planning-reasoning"],"confidence":0.5,"matches":0,"success_rate":0},{"id":"github-jnmetacode--agency-agents-zh__cap_9","uri":"capability://memory.knowledge.mcp.memory.integration.for.persistent.agent.context","name":"mcp memory integration for persistent agent context","description":"Integrates with the Model Context Protocol (MCP) to enable agents to maintain persistent memory across sessions. Agents can store and retrieve context (e.g., project state, user preferences, conversation history) via MCP memory servers, allowing them to pick up where they left off in subsequent sessions. The integration is optional and tool-dependent — agents can function without MCP, but with MCP they gain stateful capabilities. This is particularly useful for long-running projects or multi-session workflows where context preservation is critical.","intents":["Maintain agent context across multiple sessions without manual context re-entry","Enable agents to remember project state, user preferences, and conversation history","Support long-running projects that span multiple days or weeks","Reduce cognitive load on users by having agents remember previous decisions and context"],"best_for":["Teams working on long-running projects that span multiple sessions","Users who want agents to remember preferences and project context","Developers building stateful multi-agent systems","Organizations with complex workflows that require context preservation"],"limitations":["MCP integration is optional and tool-dependent — not all tools support MCP","Memory storage is external to the agent — requires a separate MCP server (e.g., local file system, database)","No built-in memory management — agents may accumulate stale or irrelevant context over time","Memory access is synchronous — may add latency to agent responses if the MCP server is slow","No privacy guarantees — memory is stored in plaintext unless the MCP server implements encryption","Memory is not shared across agents by default — requires explicit configuration"],"requires":["MCP server implementation (e.g., local file system, database, or cloud service)","Tool support for MCP (e.g., Claude Code, some versions of Cursor)","Configuration to connect agents to the MCP server"],"input_types":["Agent context (project state, user preferences, conversation history)","MCP memory server configuration (connection details, authentication)"],"output_types":["Stored context in MCP memory server","Retrieved context for agent use in subsequent sessions"],"categories":["memory-knowledge","tool-use-integration"],"confidence":0.5,"matches":0,"success_rate":0}],"trust":{"score":53,"verified":false,"data_access_risk":"high","permissions":["Bash 4.0+ or PowerShell 5.0+","At least one supported AI tool installed (Cursor, Claude Code, GitHub Copilot, Windsurf, Aider, etc.)","Read/write permissions to tool configuration directories","Git (optional, for version control of agent definitions)","Access to the GitHub repository (public, no authentication required)","Basic understanding of Markdown and YAML syntax (for reading/editing agent definitions)","One of the 14+ supported AI tools (Cursor, Claude Code, Copilot, Windsurf, etc.)","OpenClaw installed and configured","Integration configuration (which agents to deploy, target tools)","Conversion pipeline (convert.sh/install.sh) to generate tool-specific formats"],"failure_modes":["Conversion fidelity depends on target tool's schema — some advanced features may not translate (e.g., Claude Code's native memory features may not map to Cursor rules)","Bash pipeline requires Unix-like environment; PowerShell variant adds Windows support but requires separate maintenance","No real-time sync — agents must be re-deployed after updates; no hot-reload capability","Tool detection is filesystem-based (checking ~/.cursor/, ~/.claude/, etc.) and may fail with non-standard installations","Library is static — agents are not dynamically updated based on user feedback or performance metrics","Chinese market agents may not generalize well to non-Chinese contexts (e.g., Xiaohongshu strategies don't apply to TikTok)","No built-in mechanism to customize agents per-organization (e.g., company-specific tone, policies) — requires forking or manual editing","Agents are persona-based, not task-based — may require chaining multiple agents for complex workflows","OpenClaw integration is optional and requires OpenClaw to be installed and configured","OpenClaw may not support all tools in the agency-agents-zh ecosystem","builder identity is not verified yet","no observed match outcomes yet"],"rank_breakdown":{"adoption":0.6769090568478571,"quality":0.5,"ecosystem":0.8,"match_graph":0.25,"freshness":0.75,"weights":{"adoption":0.25,"quality":0.25,"ecosystem":0.1,"match_graph":0.28,"freshness":0.12}},"observed_outcomes":{"matches":0,"success_rate":0,"avg_confidence":0,"top_intents":[],"last_matched_at":null},"maintenance":{"status":"active","updated_at":"2026-05-24T12:16:21.550Z","last_scraped_at":"2026-05-03T13:59:50.673Z","last_commit":"2026-05-02T17:20:50Z"},"community":{"stars":9541,"forks":1814,"weekly_downloads":null,"model_downloads":null,"model_likes":null}},"distribution":{"claim_url":"https://unfragile.ai/submit?claim=jnmetacode--agency-agents-zh","compare_url":"https://unfragile.ai/compare?artifact=jnmetacode--agency-agents-zh"}},"signature":"M4dywT20UfmaQWm3RwAx7KM+AsR2xx1b6k4Uh+/dBcQNLRwJrjrmPHiy3RTX2uXIu+KNvpwDRz0f4leGx1a+CA==","signedAt":"2026-06-21T08:54:12.731Z","signedBy":"unfragile.ai","version":1},"_links":{"self":"https://unfragile.ai/api/v1/passport/jnmetacode--agency-agents-zh","artifact":"https://unfragile.ai/jnmetacode--agency-agents-zh","verify":"https://unfragile.ai/api/v1/verify?slug=jnmetacode--agency-agents-zh","publicKey":"https://unfragile.ai/api/v1/trust-passport-public-key","spec":"https://unfragile.ai/trust","schema":"https://unfragile.ai/schema.json","docs":"https://unfragile.ai/docs"}}