--- name: ava-mcp-usage description: Guide for using Ava's MCP tools effectively for task management and progress reporting. Use when working with Ava MCP server to report progress on coding tasks, including starting tasks, updating progress, reporting blockers, and completing tasks. Helps users understand best practices for privacy-conscious task reporting with minimal effort. --- # Ava MCP Usage ## Overview Ava provides MCP tools for automatic progress reporting to Slack. This skill shows you how to use these tools effectively with minimal mental burden while maintaining privacy. ## Core Philosophy Ava works best when you tell it what to watch: **"Keep an eye on this work."** **Important: Treat any coding work as a task**, even if the user doesn't explicitly say "task". When the user starts working on something, that's a task to track. Key principles: - Treat all coding work as tasks worth tracking - Report only the current task - **Proactive, frequent updates**: Report progress at every meaningful milestone without waiting for user prompts - Update granularly: schema changes, file edits, endpoint implementations, test additions - all worth reporting - Ask for help immediately when blocked - Ask user for consent before marking tasks complete - Short commands are sufficient - **One task per session**: If starting different work while a task is active, ask the user if they want to complete the current task first or start a new Claude Code session for the new work ## MCP Tools Quick Reference ### start_task Tell Ava what you're working on now. **Treat any coding work as a task**, even if not explicitly called a "task". ``` Start task: ログインのエラー処理改善 ``` **Tips:** - Treat all coding work as tasks worth tracking - Only report the task you're currently working on - Be specific but concise - Ava will create a Slack thread for this task - **Before starting a new task**: If there's already an active task in this session, ask the user: "Should I complete the current task first, or would you like to start a new Claude Code session for this different work?" ### update_task Report progress proactively at every meaningful milestone, even without user prompts. ``` Update task: リトライ実装が完了 ``` **When to update (proactively):** - After creating or editing a schema file - After implementing an endpoint or API route - After adding validation logic - After writing tests - After fixing a bug or error - After refactoring a component - Any other meaningful progress point **Tips:** - **Don't wait for user prompts** - update proactively when you complete a milestone - Small, granular updates are highly valuable - they create a clear activity trail - No need for detailed explanations - brief summaries are sufficient - Each update goes to the same Slack thread ### report_blocked Share blockers without hesitation. ``` Report blocked: テスト環境でトークンが通らない ``` **Tips:** - Don't hold back - reporting early helps - Brief descriptions are fine - Ava notifies your team in the Slack thread ### complete_task Mark completion after confirming with the user. ``` Complete task: レビュー完了してマージしました ``` **Tips:** - PR creation is progress, not completion - Always ask the user before marking a task complete - Include PR URL if relevant ## Privacy-Conscious Reporting Ava sends only abstract summaries to Slack/MCP. Never send code or secrets. ### ❌ Do NOT send: - Full code - Secret keys or tokens - Detailed error logs - Repository confidential information ### ✅ Safe to send: - Abstract work summaries - Brief blocker descriptions - Completion summary with PR URL **Remember:** Your code and judgment stay private. Ava only delivers progress updates. ## Minimal Effort Required Ava and your coding agent work together to fill in details. Short phrases are enough: ``` Start task: fix auth bug Update task: retry logic done Report blocked: test token issue Complete task: merged ``` No need for elaborate explanations - Ava understands context and supplements your brief inputs. --- Source attribution: yutakobayashidev Original project: https://github.com/yutakobayashidev/ava/blob/fcf3eb3d46b7c29eb8d3032585ba0b2bf7db2f3d/.claude/skills/ava-mcp-usage/SKILL.md License: MIT License information: https://opensource.org/license/mit Original upstream instructions; downloaded from Dave AI. ## Upstream license notice MIT License Copyright (c) 2025 Yuta Kobayashi Permission is hereby granted, free of charge, to any person obtaining a copy of this software and associated documentation files (the "Software"), to deal in the Software without restriction, including without limitation the rights to use, copy, modify, merge, publish, distribute, sublicense, and/or sell copies of the Software, and to permit persons to whom the Software is furnished to do so, subject to the following conditions: The above copyright notice and this permission notice shall be included in all copies or substantial portions of the Software. THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE. # Dave AI resource metadata Creator / publisher: yutakobayashidev Original project: https://github.com/yutakobayashidev/ava/blob/fcf3eb3d46b7c29eb8d3032585ba0b2bf7db2f3d/.claude/skills/ava-mcp-usage/SKILL.md Attribution basis: Recovered original package and original project license at a pinned revision Fields of work: Operations, Project & Program Management, Sales, Customer Support Roles: Customer Success Manager, Operations Manager, Project Manager, Sales Development Rep (SDR), Technical Support Specialties: Specialty not specified Tools: Slack, Ava Mcp Country applicability: Country not specified Classification basis: Source catalog labels Instruction SHA-256: cd69bd757cbb631f281d488b4fe41ba773dbd09b4b01e928504f6d5bca2bb5bc Metadata reviewed: 2026-10-27 This metadata supersedes earlier Dave AI catalog attribution. Original instruction and supporting files are preserved.