### Install Command Example Queries Source: https://github.com/agentscope-ai/agentteams/blob/main/manager/agent/worker-agent/skills/find-skills/SKILL.md Example install commands for specific skills mentioned in the documentation. ```bash "${FIND_SKILLS_SCRIPT}" install remotion-best-practices ``` ```bash "${FIND_SKILLS_SCRIPT}" install github-operations ``` -------------------------------- ### Quick Start Installation on macOS/Linux Source: https://github.com/agentscope-ai/agentteams/blob/main/install/README.md One-click installation script for macOS/Linux using curl and bash. Requires Docker Desktop or Docker Engine to be installed and running. ```bash bash <(curl -fsSL https://raw.githubusercontent.com/agentscope-ai/AgentTeams/main/install/agentteams-install.sh) ``` -------------------------------- ### Example find command output Source: https://github.com/agentscope-ai/agentteams/blob/main/manager/agent/hermes-worker-agent/skills/find-skills/SKILL.md Example output from the find command. It shows the exact install command to use and a brief description of the skill. ```text Install with ${HERMES_HOME}/skills/find-skills/scripts/agentteams-find-skill.sh install vercel-react-best-practices └ React and Next.js performance guidance ``` -------------------------------- ### Setup Examples: Built-in Template and User-Provided YAML Source: https://github.com/agentscope-ai/agentteams/blob/main/manager/agent/skills/mcp-server-management/references/create-update-server.md Two usage examples. The first uses a built-in template for GitHub, automatically extracting the domain. The second uses a user-provided YAML and an explicit API domain for a custom service. ```bash bash .../setup-mcp-server.sh github "ghp_xxxxxxxxxxxx" ``` ```bash bash .../setup-mcp-server.sh weather "my-key" \ --yaml-file /tmp/mcp-weather.yaml --api-domain "api.weather.com" ``` -------------------------------- ### Quick Start: Clone, Configure, and Test AgentTeams Source: https://github.com/agentscope-ai/agentteams/blob/main/tests/skills/agentteams-test/SKILL.md Quick start workflow: clone the repository, create a configuration file from the example, and run the full test suite. Requires setting AGENTTEAMS_LLM_API_KEY and other environment variables in ~/agentteams-manager.env before running make test. ```bash git clone https://github.com/alibaba/agentteams.git && cd agentteams # 2. Create config file (first time) cp agentteams-manager.env.example ~/agentteams-manager.env # Edit ~/agentteams-manager.env and set AGENTTEAMS_LLM_API_KEY, etc. # 3. Run full test set -a && . ~/agentteams-manager.env && set +a && make test ``` -------------------------------- ### Example skill presentation response Source: https://github.com/agentscope-ai/agentteams/blob/main/manager/agent/hermes-worker-agent/skills/find-skills/SKILL.md Example response to present to the user when a relevant skill is found. It includes the skill name, description, and the exact install command. ```text I found a skill that might help! The "remotion-best-practices" skill provides best practices for Remotion video creation in React. To install it: "${FIND_SKILLS_SCRIPT}" install remotion-best-practices ``` -------------------------------- ### Example: weather MCP server YAML and setup command Source: https://github.com/agentscope-ai/agentteams/blob/main/manager/agent/skills/mcp-server-management/references/custom-yaml-guide.md Complete example for adding a weather API. Creates a YAML file with a get_weather tool using X-API-Key header, then runs setup-mcp-server.sh with the YAML file. The script substitutes the accessToken value. ```bash cat > /tmp/mcp-weather.yaml << 'YAML' server: name: weather-mcp-server config: accessToken: "" tools: - name: get_weather description: "Get current weather for a city" args: - name: city description: "City name (e.g., London, Tokyo)" type: string required: true - name: units description: "Temperature units" type: string required: false default: "metric" requestTemplate: url: "https://api.openweather.com/v1/weather?q={{.args.city}}&units={{.args.units}}" method: GET headers: - key: X-API-Key value: "{{.config.accessToken}}" YAML bash .../setup-mcp-server.sh weather "" --yaml-file /tmp/mcp-weather.yaml ``` -------------------------------- ### Run Full Test Suite with Make Source: https://github.com/agentscope-ai/agentteams/blob/main/tests/skills/agentteams-test/SKILL.md Loads the environment configuration and runs the full test suite, which automatically handles install, test, and uninstall phases. Requires prior setup of ~/agentteams-manager.env. ```bash # Load config and run tests (automatically executes install → test → uninstall) set -a && . ~/agentteams-manager.env && set +a && make test ``` -------------------------------- ### Quick Install with make install Source: https://github.com/agentscope-ai/agentteams/blob/main/docs/usage/development.md Builds Manager and Worker images, runs the install script, mounts the container runtime socket, and saves configuration to ./agentteams-manager.env. Only AGENTTEAMS_LLM_API_KEY is required. ```bash AGENTTEAMS_LLM_API_KEY="sk-xxx" make install ``` -------------------------------- ### Install adapter into QwenPaw working directory Source: https://github.com/agentscope-ai/agentteams/blob/main/plugins/teamharness/adapters/qwenpaw/README.md Run this script to install the adapter into the current QwenPaw working directory. It builds the package, unpacks it, and runs the qwenpaw plugin install command. ```bash plugins/teamharness/adapters/qwenpaw/install.sh ``` -------------------------------- ### Home directory sharing usage examples Source: https://github.com/agentscope-ai/agentteams/blob/main/docs/usage/manager-guide.md Demonstrates four use cases of the optional home directory sharing feature: installing with the recommended environment variable, creating files in the home directory, accessing files via the symlink path, and referencing files in agent configuration using the same path. The symlink ensures the container sees the same host path, enabling seamless file access. ```bash # Example 1: Install with home directory sharing (recommended) AGENTTEAMS_LLM_API_KEY=your-key-here ./install/agentteams-install.sh manager # Example 2: Place files in home directory for agent access mkdir -p ~/project-inputs/ echo "Sample data" > ~/project-inputs/sample.txt # Example 3: Agent can access files at the same path in container as on host # Host path: /home/zhangty/project-inputs/sample.txt # Container path: /home/zhangty/project-inputs/sample.txt (via symlink) # Example 4: Use in agent configuration to access host files # In agent configuration, refer to files using the same path as host: # Host: /home/zhangty/data/input.txt # Container: /home/zhangty/data/input.txt (identical path via symlink) ``` -------------------------------- ### QwenPaw plugin install command Source: https://github.com/agentscope-ai/agentteams/blob/main/plugins/teamharness/adapters/qwenpaw/README.md This command is run by install.sh to install the generated package. QwenPaw writes the installed plugin under ${QWENPAW_WORKING_DIR}/plugins. ```bash qwenpaw plugin install --force ``` -------------------------------- ### Install migration skill (OpenClaw prompt) Source: https://github.com/agentscope-ai/agentteams/blob/main/docs/usage/import-worker.md Asks the OpenClaw to install the agentteams-migrate skill from a given path. ```text Install the agentteams-migrate skill from /path/to/agentteams/migrate/skill/ ``` -------------------------------- ### Find Command Example Queries Source: https://github.com/agentscope-ai/agentteams/blob/main/manager/agent/worker-agent/skills/find-skills/SKILL.md Example invocations of the find command for different user requests. These are shown as bullet examples in the documentation. ```bash "${FIND_SKILLS_SCRIPT}" find react performance ``` ```bash "${FIND_SKILLS_SCRIPT}" find pr review ``` ```bash "${FIND_SKILLS_SCRIPT}" find changelog ``` -------------------------------- ### Install AgentTeams on Kubernetes with Qwen (通义千问) Source: https://github.com/agentscope-ai/agentteams/blob/main/README.md Install AgentTeams on Kubernetes with Helm using Qwen (通义千问) as the LLM provider. Sets the provider to qwen and the default model to qwen3.5-plus. Requires the same Helm and Kubernetes prerequisites as the other Helm install commands. ```bash helm install agentteams higress.io/agentteams \ -n agentteams-system --create-namespace \ --render-subchart-notes \ --set credentials.llmApiKey= \ --set credentials.llmProvider=qwen \ --set credentials.defaultModel=qwen3.5-plus \ --set credentials.adminPassword= \ --set gateway.publicURL=http://localhost:18080 ``` -------------------------------- ### Install skill directly from market Source: https://github.com/agentscope-ai/agentteams/blob/main/manager/agent/hermes-worker-agent/skills/find-skills/SKILL.md Installs a skill directly from the market when the user already knows the skill name. The skill name is specified in the placeholder. ```bash "${FIND_SKILLS_SCRIPT}" install ``` -------------------------------- ### GET /api/v1/projects - 200 OK JSON response Source: https://github.com/agentscope-ai/agentteams/blob/main/docs/usage/project-workflow-api.md Example response for listing projects. Shows project fields including project_id, title, status, plan_type, team_id, and mode. ```json { "projects": [ { "project_id": "demo-project-001", "title": "Demo project", "status": "active", "plan_type": "dag", "team_id": "biz-team", "mode": "project" } ], "total": 1 } ``` -------------------------------- ### Install DingTalk plugin locally Source: https://github.com/agentscope-ai/agentteams/blob/main/docs/usage/integrations/dingtalk.md Installs the DingTalk plugin from the local directory. Run this after cloning and installing dependencies. ```bash openclaw plugins install -l . ``` -------------------------------- ### Individual Install, Uninstall, and Test with Existing Installation Source: https://github.com/agentscope-ai/agentteams/blob/main/tests/skills/agentteams-test/SKILL.md Provides individual commands for installation only, uninstallation only, and running tests using an existing installation. The install step uses AGENTTEAMS_YOLO=1 to skip confirmation; the test runner uses --skip-build and --use-existing to avoid rebuilding. ```bash # Install only set -a && . ~/agentteams-manager.env && set +a && AGENTTEAMS_YOLO=1 make install # Uninstall only make uninstall # Run tests using existing installation (skip reinstall) set -a && . ~/agentteams-manager.env && set +a ./tests/run-all-tests.sh --skip-build --use-existing ``` -------------------------------- ### Prepare Git Request Example Source: https://github.com/agentscope-ai/agentteams/blob/main/manager/agent/worker-skills/git-delegation/SKILL.md Example of a git-request message with a workspace and operations list for cloning and branching. ```text @manager:DOMAIN task-20260225 git-request: workspace: /root/agentteams-fs/shared/tasks/task-20260225/workspace operations: - git clone https://github.com/agentscope-ai/AgentTeams.git - cd agentteams && git checkout -b feature-xyz ---CONTEXT--- Starting work on feature XYZ ---END--- ``` -------------------------------- ### No auth example: setup-mcp-proxy.sh sentry Source: https://github.com/agentscope-ai/agentteams/blob/main/manager/agent/skills/mcp-server-management/references/setup-mcp-proxy.md Example with no auth header. Uses http transport and a sample Sentry MCP server URL. ```bash # No auth bash .../setup-mcp-proxy.sh sentry https://mcp.sentry.dev/mcp http ``` -------------------------------- ### Example docker ps output Source: https://github.com/agentscope-ai/agentteams/blob/main/docs/usage/integrations/dingtalk.md Example output from 'docker ps -a' showing the Worker container. In this example, run 'docker restart agentteams-worker-fbi-claw' to restart the Worker container. ```text 799c1ca06455