{"passport":{"unfragile":{"@version":"1.0","version":"2026-05","artifact":{"id":"github-ufomiao--zcf","slug":"ufomiao--zcf","name":"zcf","type":"agent","url":"http://zcf.ufomiao.com/","page_url":"https://unfragile.ai/ufomiao--zcf","categories":["automation"],"tags":["agent","ai","ai-agent","bmad-method","ccr","claude","claude-4","claude-ai","claude-code","cli","gpt","gpt-5","llm","llm-code","nodejs","openai","prompt","typescript","workflow","zcf"],"pricing":{"model":"open_source","free":true,"starting_price":null},"status":"active","verified":false},"capabilities":[{"id":"github-ufomiao--zcf__cap_0","uri":"capability://automation.workflow.unified.dual.tool.configuration.abstraction","name":"unified dual-tool configuration abstraction","description":"Provides a single CLI interface that abstracts configuration complexity for two distinct AI coding tools (Claude Code VS Code extension and Codex terminal CLI) through tool-specific adapter pattern. Uses TOML-based configuration files with tool-specific managers that translate unified settings into tool-native formats, eliminating the need for users to manually configure each tool separately. Implements automatic platform detection and intelligent defaults to minimize required user input.","intents":["I want to configure both Claude Code and Codex without learning each tool's separate configuration system","I need to switch between different API providers across both tools instantly","I want to maintain consistent settings across multiple AI coding environments"],"best_for":["developers using both Claude Code and Codex simultaneously","teams standardizing on multiple AI coding tools","users managing complex multi-provider setups"],"limitations":["Abstraction layer adds complexity if users need tool-specific advanced features not exposed by ZCF","Configuration changes require CLI invocation rather than direct file editing for consistency","Limited to the two supported tools; cannot manage other AI coding environments"],"requires":["Node.js 14+","npm or pnpm package manager","Claude Code VS Code extension installed (for Claude Code configuration)","Codex CLI installed (for Codex configuration)"],"input_types":["CLI arguments","interactive prompts","TOML configuration files","environment variables"],"output_types":["TOML configuration files","tool-specific configuration formats","CLI status messages"],"categories":["automation-workflow","tool-use-integration"],"confidence":0.5,"matches":0,"success_rate":0},{"id":"github-ufomiao--zcf__cap_1","uri":"capability://automation.workflow.multi.profile.api.provider.switching.with.presets","name":"multi-profile api provider switching with presets","description":"Manages multiple API provider configurations (OpenAI, Anthropic Claude, etc.) with instant switching capability through a preset system stored in TOML files. Users define named profiles containing API keys, model selections, and provider-specific settings, then switch between them via CLI commands without reconfiguring. The system validates API credentials and maintains provider-specific defaults for each tool adapter.","intents":["I want to test code generation across different LLM providers without manual reconfiguration","I need to switch between production and development API keys instantly","I want to maintain separate billing/quota profiles for different projects"],"best_for":["developers evaluating multiple LLM providers","teams with separate API budgets per project","researchers comparing model outputs across providers"],"limitations":["API keys stored in plaintext TOML files; requires OS-level file permissions for security","No built-in credential rotation or expiration management","Provider switching requires CLI invocation; no automatic fallback if primary provider fails"],"requires":["Valid API keys for at least one supported provider (OpenAI, Anthropic, etc.)","File system write access to configuration directory","Network connectivity to validate API credentials"],"input_types":["API keys (string)","provider names (enum)","model identifiers (string)","provider-specific settings (key-value pairs)"],"output_types":["active provider configuration (JSON/TOML)","validation status (success/error)","available profiles list (structured text)"],"categories":["automation-workflow","tool-use-integration"],"confidence":0.5,"matches":0,"success_rate":0},{"id":"github-ufomiao--zcf__cap_10","uri":"capability://automation.workflow.non.interactive.configuration.mode.for.ci.cd.automation","name":"non-interactive configuration mode for ci/cd automation","description":"Supports fully automated configuration via environment variables and command-line flags without interactive prompts, enabling ZCF integration into CI/CD pipelines and automated deployment scripts. The system reads configuration from environment variables (e.g., `ZCF_API_KEY`, `ZCF_PROVIDER`, `ZCF_LANGUAGE`) and applies them without user interaction. Non-interactive mode validates all required parameters before proceeding and fails fast with clear error messages if configuration is incomplete.","intents":["I want to configure ZCF in CI/CD pipelines without interactive prompts","I need to deploy ZCF across multiple machines with identical configurations","I want to automate ZCF setup in Docker containers or cloud environments"],"best_for":["DevOps engineers deploying ZCF at scale","CI/CD pipeline builders integrating ZCF","containerized application developers"],"limitations":["Requires all configuration parameters provided upfront; no interactive fallback","Environment variable names must be exact; typos fail silently or use defaults","No validation of API credentials until first use; configuration errors discovered late"],"requires":["All required environment variables set before ZCF invocation","Valid API keys and provider names in environment","Understanding of ZCF environment variable schema"],"input_types":["environment variables (ZCF_* prefix)","command-line flags (--non-interactive, --api-key, etc.)","configuration files (TOML, pre-created)"],"output_types":["configuration files (TOML)","validation status (exit code)","error messages (stderr)"],"categories":["automation-workflow"],"confidence":0.5,"matches":0,"success_rate":0},{"id":"github-ufomiao--zcf__cap_11","uri":"capability://automation.workflow.uninstall.and.cleanup.system","name":"uninstall and cleanup system","description":"Provides complete uninstallation capability that removes ZCF package, configuration files, and backup history with optional preservation of user data. The `uninstall` command removes npm package, deletes configuration directories, and cleans up any created symlinks or PATH modifications. Users can choose to preserve configurations for later restoration or completely remove all traces of ZCF from their system.","intents":["I want to completely remove ZCF from my system","I need to uninstall ZCF but preserve my configurations for later use","I want to clean up ZCF installation without manual file deletion"],"best_for":["users discontinuing ZCF usage","developers testing ZCF installation/uninstallation","teams managing software lifecycle"],"limitations":["Uninstall cannot remove npm global package if installed with sudo; requires same permissions as installation","Preserved configurations remain on disk; user must manually delete if complete cleanup desired","No rollback after uninstall; configurations can be restored but not automatically"],"requires":["Write permissions to ZCF installation directory","Write permissions to configuration directory"],"input_types":["uninstall command with optional flags (--keep-config, --force)"],"output_types":["uninstall status messages","list of removed files/directories","preserved configuration location (if applicable)"],"categories":["automation-workflow"],"confidence":0.5,"matches":0,"success_rate":0},{"id":"github-ufomiao--zcf__cap_12","uri":"capability://data.processing.analysis.toml.based.configuration.file.management.with.schema.validation","name":"toml-based configuration file management with schema validation","description":"Uses TOML format for all configuration files with structured schema defining valid keys, types, and constraints. The system validates configuration files against schema on load, providing clear error messages for invalid configurations. Configuration is organized hierarchically (global ZCF config, tool-specific configs, workflow configs) with inheritance and override mechanisms. The system supports configuration comments and provides default values for optional keys.","intents":["I want to manually edit configuration files with clear structure and validation","I need to understand what configuration options are available and valid","I want configuration changes to be validated before taking effect"],"best_for":["developers comfortable editing configuration files","teams maintaining configuration in version control","users wanting to understand ZCF configuration structure"],"limitations":["TOML syntax errors cause configuration load failure; no graceful degradation","Schema validation is strict; typos in configuration keys silently use defaults","Manual TOML editing bypasses ZCF's validation; users can create invalid configurations"],"requires":["TOML parser library (included in ZCF dependencies)","Text editor for manual configuration editing","Understanding of TOML syntax"],"input_types":["TOML configuration files","configuration key-value pairs (via CLI or API)"],"output_types":["validated configuration objects (JavaScript)","configuration validation errors","configuration file contents (TOML)"],"categories":["data-processing-analysis","automation-workflow"],"confidence":0.5,"matches":0,"success_rate":0},{"id":"github-ufomiao--zcf__cap_13","uri":"capability://code.generation.editing.language.and.model.configuration.per.tool","name":"language and model configuration per tool","description":"Allows per-tool configuration of programming language support and AI model selection, with language-specific defaults and model-specific parameters. Users can specify which programming languages each tool should support, set default models for different task types, and configure language-specific prompts and output formatting. The system maintains language-to-model mappings and validates that selected models are available from configured API providers.","intents":["I want Claude Code to focus on Python while Codex handles JavaScript","I need different models for different programming languages","I want language-specific output formatting (e.g., docstring styles)"],"best_for":["polyglot development teams with language specialization","developers optimizing model selection per language","teams with language-specific coding standards"],"limitations":["Language-specific configuration is per-tool; no global language preferences","Model availability depends on API provider; not all models support all languages","Language detection is manual; no automatic language detection from code context"],"requires":["API provider configured with model availability information","Valid model identifiers for selected provider"],"input_types":["language names (Python, JavaScript, etc.)","model identifiers (gpt-4, claude-3-opus, etc.)","language-specific parameters (TOML)"],"output_types":["language-model mappings (JSON)","language-specific configuration (TOML)","validation results"],"categories":["code-generation-editing","automation-workflow"],"confidence":0.5,"matches":0,"success_rate":0},{"id":"github-ufomiao--zcf__cap_2","uri":"capability://automation.workflow.zero.configuration.initialization.with.platform.detection","name":"zero-configuration initialization with platform detection","description":"Automatically detects the user's operating system (Windows, macOS, Linux, Termux, WSL) and installs ZCF with platform-appropriate defaults and paths. The `init` command performs one-time setup including dependency validation, configuration directory creation, and interactive prompts for essential settings (API keys, preferred language, default models). Uses environment variable detection and file system checks to infer user preferences and minimize required input.","intents":["I want to set up ZCF quickly without understanding its architecture","I need ZCF to work on my specific OS without manual path configuration","I want sensible defaults that I can customize later if needed"],"best_for":["first-time users of ZCF","developers new to AI coding tools","teams deploying ZCF across heterogeneous OS environments"],"limitations":["Interactive prompts cannot be fully automated in CI/CD pipelines; requires `--non-interactive` flag with environment variables","Platform detection may fail on exotic OS variants or heavily customized environments","Initial setup requires internet connectivity to validate API credentials"],"requires":["Node.js 14+ installed and in PATH","npm or pnpm package manager","Write access to home directory for configuration storage","At least one valid API key (OpenAI or Anthropic recommended)"],"input_types":["interactive CLI prompts","environment variables (for non-interactive mode)","command-line flags"],"output_types":["configuration files (TOML)","initialization status messages","setup validation results"],"categories":["automation-workflow"],"confidence":0.5,"matches":0,"success_rate":0},{"id":"github-ufomiao--zcf__cap_3","uri":"capability://planning.reasoning.workflow.template.system.with.customizable.output.styles","name":"workflow template system with customizable output styles","description":"Provides pre-built workflow templates (SixStep Workflow, Git Workflow, BMad Enterprise Workflow) that define multi-step AI coding processes with customizable output styles and AI personalities. Templates are stored as configuration files that specify prompt sequences, tool invocations, and output formatting rules. Users can create custom workflows by extending template structure, and output styles control how AI responses are formatted (tone, detail level, structure). The system uses i18next for internationalization of workflow prompts and output styles.","intents":["I want to run standardized multi-step coding workflows without manually orchestrating each step","I need AI responses formatted in a specific style (e.g., detailed technical vs. concise)","I want to create team-standard workflows that enforce consistent AI coding practices"],"best_for":["teams with standardized development processes","enterprises enforcing coding standards via AI","developers wanting to automate complex multi-step coding tasks"],"limitations":["Workflow execution is sequential; no parallel step execution or conditional branching","Output style customization limited to prompt-level control; cannot modify AI model behavior directly","Custom workflows require understanding ZCF's template schema; no visual workflow builder"],"requires":["ZCF initialized with at least one API provider configured","TOML knowledge for creating custom workflows","Understanding of target tool's capabilities (Claude Code or Codex)"],"input_types":["workflow template definitions (TOML)","output style configurations (TOML)","user code context (files, directories)","workflow parameters (CLI arguments)"],"output_types":["multi-step AI responses (formatted per output style)","code modifications (if applicable)","workflow execution logs"],"categories":["planning-reasoning","automation-workflow"],"confidence":0.5,"matches":0,"success_rate":0},{"id":"github-ufomiao--zcf__cap_4","uri":"capability://tool.use.integration.mcp.service.one.click.configuration.and.integration","name":"mcp service one-click configuration and integration","description":"Provides automated setup for Model Context Protocol (MCP) services through a configuration interface that handles service discovery, credential management, and tool-specific integration. Users select desired MCP services from a menu, and ZCF automatically configures them for both Claude Code and Codex with minimal manual intervention. The system manages MCP service credentials, validates service connectivity, and exposes MCP capabilities to the underlying AI tools through standardized integration points.","intents":["I want to add MCP services to my AI coding environment without manual configuration","I need MCP services available in both Claude Code and Codex simultaneously","I want to validate that MCP services are properly connected before using them"],"best_for":["developers using MCP-compatible services (databases, APIs, file systems)","teams standardizing on MCP for tool integration","users wanting to extend AI tool capabilities with external services"],"limitations":["Limited to MCP services that ZCF has pre-configured; custom MCP services require manual setup","MCP service credentials stored in same TOML files as API keys; no separate credential vault","Service validation is one-time at configuration; no continuous health monitoring"],"requires":["ZCF initialized with API provider configured","MCP server binaries or Docker containers for selected services","Network access to MCP service endpoints","Appropriate credentials for each MCP service"],"input_types":["MCP service selection (interactive menu)","service credentials (API keys, connection strings)","service-specific configuration parameters"],"output_types":["MCP service configuration (TOML)","service validation results","integration status for each tool"],"categories":["tool-use-integration","automation-workflow"],"confidence":0.5,"matches":0,"success_rate":0},{"id":"github-ufomiao--zcf__cap_5","uri":"capability://automation.workflow.version.management.and.automatic.update.detection","name":"version management and automatic update detection","description":"Monitors ZCF package version against npm registry and notifies users of available updates through the `check-updates` command. Implements semantic versioning validation and provides `update` command for in-place package upgrades with automatic configuration migration. The system maintains version compatibility checks for dependent tools (Claude Code, Codex) and validates configuration file compatibility across versions.","intents":["I want to know when ZCF updates are available without manually checking npm","I need to upgrade ZCF while preserving my existing configurations","I want to ensure my configurations remain compatible after updating"],"best_for":["developers wanting to stay current with ZCF improvements","teams managing ZCF across multiple machines","users concerned about security updates"],"limitations":["Update detection requires npm registry connectivity; fails silently if offline","Automatic update may fail if npm permissions are insufficient","Configuration migration only handles schema changes; custom workflows may require manual updates"],"requires":["npm or pnpm package manager with registry access","Internet connectivity to check npm registry","Write permissions to ZCF installation directory"],"input_types":["CLI command invocation","version comparison logic"],"output_types":["version information (current, latest, available)","update availability status","migration logs (if applicable)"],"categories":["automation-workflow"],"confidence":0.5,"matches":0,"success_rate":0},{"id":"github-ufomiao--zcf__cap_6","uri":"capability://text.generation.language.multi.language.internationalization.with.i18next","name":"multi-language internationalization with i18next","description":"Provides full internationalization support for English, Chinese (Simplified), and Japanese through i18next framework integration. All CLI prompts, error messages, workflow templates, and output styles are translatable via JSON translation files organized by language code. The system auto-detects user locale from OS settings or allows explicit language selection via configuration, then loads appropriate translation files at runtime for all user-facing text.","intents":["I want to use ZCF in my native language without English prompts","I need to deploy ZCF to teams across different language regions","I want to contribute translations for additional languages"],"best_for":["non-English speaking developers","international teams with language diversity","organizations standardizing on specific languages"],"limitations":["Only three languages currently supported (English, Chinese Simplified, Japanese); adding new languages requires code contribution","Translation files must be manually maintained; no automatic translation service","Some technical terms may not translate well; fallback to English for untranslated keys"],"requires":["i18next library installed (included in ZCF dependencies)","Translation files present in package (included in distribution)"],"input_types":["language code selection (en, zh-CN, ja-JP)","OS locale detection"],"output_types":["localized CLI prompts and messages","localized workflow templates","localized error messages"],"categories":["text-generation-language","automation-workflow"],"confidence":0.5,"matches":0,"success_rate":0},{"id":"github-ufomiao--zcf__cap_7","uri":"capability://automation.workflow.backup.and.recovery.system.for.configurations","name":"backup and recovery system for configurations","description":"Automatically creates timestamped backups of configuration files before modifications and provides recovery mechanism to restore previous configurations. The system maintains a backup history in a dedicated directory with metadata about each backup (timestamp, change description, version). Users can list available backups and restore to any previous state via CLI commands, with validation to ensure restored configurations are compatible with current ZCF version.","intents":["I accidentally changed my configuration and want to revert to a previous state","I want to maintain a history of configuration changes for audit purposes","I need to restore configurations after a failed update"],"best_for":["users making frequent configuration changes","teams requiring configuration audit trails","developers testing different configurations"],"limitations":["Backup storage consumes disk space; no automatic cleanup of old backups","Recovery is manual via CLI; no automatic rollback on configuration errors","Backups are local only; no cloud backup or synchronization"],"requires":["File system write access to backup directory","Sufficient disk space for backup storage"],"input_types":["configuration file modifications","backup restore commands","backup selection (timestamp or index)"],"output_types":["backup files (TOML, timestamped)","backup metadata (JSON)","recovery status messages"],"categories":["automation-workflow"],"confidence":0.5,"matches":0,"success_rate":0},{"id":"github-ufomiao--zcf__cap_8","uri":"capability://automation.workflow.cli.command.routing.and.interactive.menu.system","name":"cli command routing and interactive menu system","description":"Implements a command-based CLI interface with both direct command invocation and interactive menu navigation. The router dispatches commands to appropriate handlers based on command name, with support for both interactive prompts and non-interactive flag-based invocation. The menu system provides hierarchical navigation through available commands with descriptions, enabling discoverability for users unfamiliar with CLI options. Commands include `init`, `update`, `menu`, `config-switch`, `check-updates`, `uninstall`, `ccr`, and `ccu`.","intents":["I want to navigate ZCF commands through an interactive menu without memorizing CLI syntax","I need to invoke ZCF commands programmatically with flags for automation","I want to discover available commands and their purposes"],"best_for":["CLI-novice users preferring interactive navigation","developers automating ZCF via scripts","teams with mixed technical skill levels"],"limitations":["Interactive menu requires terminal with TTY support; fails in non-interactive environments","Command routing adds latency for simple operations","Menu navigation slower than direct command invocation for experienced users"],"requires":["Terminal with TTY support (for interactive mode)","Node.js 14+ with readline module"],"input_types":["CLI command names (string)","command flags and arguments","interactive menu selections (numeric or text)"],"output_types":["command execution results","menu display (formatted text)","help text and descriptions"],"categories":["automation-workflow"],"confidence":0.5,"matches":0,"success_rate":0},{"id":"github-ufomiao--zcf__cap_9","uri":"capability://tool.use.integration.external.tool.integration.ccr.cometix.ccusage","name":"external tool integration (ccr, cometix, ccusage)","description":"Provides integration points for external tools: CCR (Claude Code Router) for intelligent tool selection, Cometix for additional code analysis, and CCUsage for API usage tracking. The system detects installed external tools and exposes their functionality through ZCF commands (`ccr` and `ccu`), with automatic credential passing and output formatting. Integration is optional; ZCF functions without external tools but gains enhanced capabilities when they are installed.","intents":["I want to automatically route code tasks to the best AI tool (Claude Code vs Codex)","I need to track API usage and costs across multiple providers","I want to analyze code quality before sending to AI tools"],"best_for":["developers using multiple AI coding tools and wanting intelligent routing","teams tracking API costs and usage","organizations with code quality requirements"],"limitations":["External tools must be installed separately; ZCF does not bundle them","Integration requires external tool compatibility with ZCF's interface","No fallback if external tool fails; error propagates to user"],"requires":["External tool binaries installed and in PATH (optional but recommended)","Appropriate credentials for external tools"],"input_types":["code context (files, directories)","task descriptions","routing criteria (for CCR)"],"output_types":["tool routing recommendations (CCR)","code analysis results (Cometix)","usage statistics (CCUsage)"],"categories":["tool-use-integration","planning-reasoning"],"confidence":0.5,"matches":0,"success_rate":0}],"trust":{"score":46,"verified":false,"data_access_risk":"high","permissions":["Node.js 14+","npm or pnpm package manager","Claude Code VS Code extension installed (for Claude Code configuration)","Codex CLI installed (for Codex configuration)","Valid API keys for at least one supported provider (OpenAI, Anthropic, etc.)","File system write access to configuration directory","Network connectivity to validate API credentials","All required environment variables set before ZCF invocation","Valid API keys and provider names in environment","Understanding of ZCF environment variable schema"],"failure_modes":["Abstraction layer adds complexity if users need tool-specific advanced features not exposed by ZCF","Configuration changes require CLI invocation rather than direct file editing for consistency","Limited to the two supported tools; cannot manage other AI coding environments","API keys stored in plaintext TOML files; requires OS-level file permissions for security","No built-in credential rotation or expiration management","Provider switching requires CLI invocation; no automatic fallback if primary provider fails","Requires all configuration parameters provided upfront; no interactive fallback","Environment variable names must be exact; typos fail silently or use defaults","No validation of API credentials until first use; configuration errors discovered late","Uninstall cannot remove npm global package if installed with sudo; requires same permissions as installation","builder identity is not verified yet","no observed match outcomes yet"],"rank_breakdown":{"adoption":0.5955950796273672,"quality":0.35,"ecosystem":0.6000000000000001,"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":"2026-05-24T12:16:22.064Z","last_scraped_at":"2026-05-03T13:57:04.027Z","last_commit":"2026-04-30T12:47:31Z"},"community":{"stars":5959,"forks":415,"weekly_downloads":null,"model_downloads":null,"model_likes":null}},"distribution":{"claim_url":"https://unfragile.ai/submit?claim=ufomiao--zcf","compare_url":"https://unfragile.ai/compare?artifact=ufomiao--zcf"}},"signature":"kF5UIix3wRjNd44YTWSY00YqdXIZsJziVintuZWtDF5dmv9l3o72p+VBWV36JQjrtRzvCKvgg5yY62U4Cv0BBA==","signedAt":"2026-06-19T21:32:13.336Z","signedBy":"unfragile.ai","version":1},"_links":{"self":"https://unfragile.ai/api/v1/passport/ufomiao--zcf","artifact":"https://unfragile.ai/ufomiao--zcf","verify":"https://unfragile.ai/api/v1/verify?slug=ufomiao--zcf","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"}}