Manage AI agent skills across Claude Code, Amp, Codex, OpenCode, Factory, and others using agents.toml. Operations teams can define skills once and deploy across multiple agents, streamlining workflows and reducing redundancy.
git clone https://github.com/803/skills-supply.gitThe skills-supply automation skill is designed to streamline the management of AI agent skills across various platforms, including Claude Code, Codex, and OpenCode. By utilizing the agents.toml configuration file, users can efficiently organize and deploy their AI agent capabilities, ensuring that their workflows are both effective and adaptable. This skill is particularly beneficial for those looking to enhance their AI automation processes without the need for extensive coding knowledge. One of the key benefits of the skills-supply skill is its potential to save time in managing AI agents. Although the exact time savings are currently unknown, the intermediate complexity of this skill means that users can expect to implement it within approximately 30 minutes. This quick setup allows developers and product managers to focus on more strategic tasks rather than getting bogged down in the intricacies of agent management. By simplifying the process, this skill facilitates smoother workflow automation, enabling teams to leverage AI effectively. This skill is particularly suited for developers, product managers, and AI practitioners who are involved in AI automation projects. Its medium GTM relevance indicates that while it may not be essential for every team, those who adopt it will likely see improvements in their operational efficiency. Practical use cases include managing multiple AI agents for different projects or ensuring that the right skills are deployed for specific tasks, which can significantly enhance productivity. With an intermediate level of difficulty, the skills-supply skill requires users to have a basic understanding of AI agent configurations. However, its implementation is straightforward, making it accessible for those familiar with AI workflows. By integrating this skill into their AI-first strategies, teams can ensure that their automation processes are not only effective but also scalable, paving the way for future innovations in AI automation.
["Step 1: Define the skill in `agents.toml`\n - Open your project’s `agents.toml` file.\n - Add the skill configuration under `[[skills]]` with platforms, description, and deployment targets. Example:\n ```toml\n [[skills]]\n name = \"skills-supply\"\n platforms = [\"claude-code\", \"amp\"]\n deploy_to = [\"agent-01\", \"agent-02\", \"agent-03\"]\n ```\n - Tip: Use comments to document the skill’s purpose and dependencies.\n\nStep 2: Validate the configuration\n - Run `claude-code validate agents.toml` (or equivalent for your platform) to check for syntax errors.\n - Tip: Fix any errors before proceeding to avoid deployment failures.\n\nStep 3: Deploy the skill\n - Execute the deployment command for your platform. Example:\n ```bash\n claude-code deploy skills-supply --force\n ```\n - Monitor the logs for errors or warnings. Tip: Use `--verbose` for detailed logs.\n\nStep 4: Integrate with Sortd\n - Map the deployed skill to Sortd’s Kanban workflows:\n 1. Open Sortd in Gmail and navigate to your team’s board (e.g., \"Sales Pipeline\").\n 2. Create a new automation rule: \"When [SKILL_TRIGGER_EVENT], create a task in [SORTD_BOARD] and assign to [AGENT].\"\n 3. Example: \"When a high-priority email is received, create a task in 'High Priority' board and assign to 'sales-assistant-01'.\"\n - Tip: Use Sortd’s API or webhooks for advanced integrations.\n\nStep 5: Test and validate\n - Perform end-to-end testing:\n 1. Trigger the skill manually or via an automated event.\n 2. Verify the task appears in Sortd under the correct board and agent.\n 3. Check logs for errors in skill execution.\n - Tip: Test edge cases (e.g., missing agents, invalid configurations).\n\nStep 6: Monitor and iterate\n - Track performance metrics in Sortd (e.g., task completion time, agent workload).\n - Use feedback from the team to refine the skill configuration or Sortd workflows.\n - Tip: Schedule a review after 1-2 weeks to assess impact and plan improvements.","Use this skill to centralize AI agent management across multiple platforms, reducing redundancy and improving workflow efficiency. Pair it with Sortd to create automated, visual task management systems directly from email communications. Ideal for operations teams, sales teams, and customer support groups that rely on Gmail and need to streamline agent deployments and task assignments.","This skill is valuable for teams that use multiple AI agents (e.g., Claude Code, Amp, Codex) and need to manage skills consistently. By defining skills once in `agents.toml`, you eliminate the need to update each agent individually, saving time and reducing errors. When integrated with Sortd, it transforms Gmail into a powerful task management system where skills can automatically generate tasks, assign follow-ups, and track progress—all without leaving your inbox. This is particularly useful for sales teams handling high volumes of leads, support teams managing customer inquiries, or operations teams coordinating complex workflows. The key advantage is the reduction of manual work and the ability to scale AI agent capabilities seamlessly across your organization."]
No install command available. Check the GitHub repository for manual installation instructions.
git clone https://github.com/803/skills-supplyCopy the install command above and run it in your terminal.
Launch Claude Code, Cursor, or your preferred AI coding agent.
Use the prompt template or examples below to test the skill.
Adapt the skill to your specific use case and workflow.
Generate a deployment plan for deploying the skill [SKILL_NAME] across [AGENT_PLATFORM] using agents.toml. Include steps for validating the skill's functionality post-deployment, and ensure compatibility with Sortd's Kanban-style email workflows for [TEAM_NAME] in [COMPANY_NAME].
### Deployment Plan for `skills-supply` Across Claude Code Agents
**Objective:** Deploy the `skills-supply` skill across 15 Claude Code agents used by the Sales Operations team at TechFlow Inc. to manage AI agent skills centrally via `agents.toml`.
#### Step 1: Skill Definition in `agents.toml`
```toml
[[skills]]
name = "skills-supply"
version = "1.2.0"
platforms = ["claude-code"]
description = "Manages AI agent skills across multiple platforms using agents.toml"
[skills.config]
deploy_to = ["sales-assistant-01", "sales-assistant-02", "sales-assistant-03", "sales-assistant-04", "sales-assistant-05",
"sales-assistant-06", "sales-assistant-07", "sales-assistant-08", "sales-assistant-09", "sales-assistant-10",
"sales-assistant-11", "sales-assistant-12", "sales-assistant-13", "sales-assistant-14", "sales-assistant-15"]
auto_update = true
```
#### Step 2: Validation of `agents.toml`
- **Tool:** `claude-code validate agents.toml`
- **Output:** Confirmed no syntax errors. All 15 agents listed under `deploy_to` are valid.
#### Step 3: Skill Deployment
- **Command:** `claude-code deploy skills-supply --force`
- **Result:** Skill deployed to all 15 agents. Logs show successful installation with no errors.
#### Step 4: Integration with Sortd
- **Action:** Mapped the skill to Sortd’s Kanban workflows for the Sales team:
- **Pipeline:** "Lead Follow-up" board in Sortd.
- **Skill Trigger:** Auto-assigns follow-up tasks to agents based on email priority (e.g., high-priority leads → `sales-assistant-01`).
- **Validation:** Tested by sending a test email to `[email protected]` with subject "URGENT: Demo Request." The task was auto-created in Sortd under "High Priority" and assigned to `sales-assistant-01`.
#### Step 5: Post-Deployment Testing
- **Test Cases:**
1. **Skill Functionality:** Verified that updating `agents.toml` auto-updates the skill across all agents.
2. **Sortd Integration:** Confirmed that tasks generated by the skill appear in the correct Sortd board and are assigned to the right agent.
3. **Error Handling:** Simulated a failed deployment by removing one agent from `deploy_to`. The system correctly flagged the missing agent and provided a remediation step.
#### Step 6: Documentation and Training
- **Documentation:** Updated the internal wiki with:
- Steps to add/remove agents from `agents.toml`.
- How to map skills to Sortd workflows.
- **Training:** Conducted a 30-minute session with the Sales Operations team to demonstrate the new workflow.
**Outcome:** The `skills-supply` skill is now centrally managed across all Claude Code agents, reducing manual updates by 90%. Integration with Sortd ensures seamless task assignment and follow-up, improving team productivity by 25% based on initial metrics.
**Next Steps:** Monitor performance for 2 weeks, then expand deployment to Amp and Codex platforms.AI assistant built for thoughtful, nuanced conversation
Get more done every day with Microsoft Teams – powered by AI
Automate security compliance and monitor real-time security posture seamlessly.
Automate your spreadsheet tasks with AI power
Agentic AI Workflow platform
Connected workspace for docs, wikis, and projects
Take a free 3-minute scan and get personalized AI skill recommendations.
Take free scan