Build and troubleshoot MCP servers, which connect AI agents to external tools, for codewhale. Covers local and remote connections, credential configuration, tool validation, and testing both successful requests and failures.
Investigate software failures by reproducing the problem, reducing the failing example, and locating its root cause. Receive an evidence-based diagnosis that separates symptoms from causes, with fixes applied only when authorized.
Prepare a specified software version for release by checking version consistency and changelog readiness, building packages, and testing basic behavior. Produce checksums, release notes, and a readiness report; public release actions require separate authorization.
Review GitHub issue discussions and code on the target branch to confirm a fix has landed, then close resolved issues with evidence, contributor credit, and an invitation to reopen if the problem returns.
Assigns a specified set of GitHub issues to a milestone, owners, or both. Checks that each target is an open issue, applies changes individually, and reports results with verification of milestone counts and selected issue details.
Create, edit, inspect, and verify PowerPoint slide decks in PPTX format. Use a consistent, simple layout and consolidate information into fewer slides, then reopen the presentation to check its slide count and titles.
Generate independent options for design, implementation, explanations, or debugging, then compare them using one shared rubric. Code candidates use separate working copies and undergo independent verification before a winner is presented for review.
Plan Codewhale changes by locating the existing implementation and its callers, defining one reviewable change, and choosing verification before editing. Use it for features, upgrades, and refactoring to avoid duplicate implementations and unfinished migrations.
Turn supplied data into a suitable chart, dashboard, or visual explanation. The workflow chooses readable ways to show comparisons, includes source and assumption notes, and checks axes, units, and missing values without inventing data or hiding uncertainty.
Manage Codewhale fleet runs by inspecting workers, the agents carrying out tasks, and reviewing logs and artifacts. Classify failures, choose a supported recovery action or draft a request for human review, and summarize evidence and follow-up ownership.
Split a large goal into clearly bounded tasks that can run in parallel. Coordinate existing agent and worktree tools, then combine and verify the results. The skill starts only when explicitly requested.
Turn an understood task into an ordered implementation plan in Codewhale. Clarify the goal and constraints, identify which steps depend on others, and record verification methods, risks, and open questions before broader implementation begins.
Record goals with measurable targets and deadlines, review progress against reported results, and update them as circumstances change. Keep a dated history in one file, with goal wording confirmed before it is saved.
Establish the current repository state before starting or resuming Codewhale development. Check the active branch, uncommitted changes, related working directories, applicable instructions, and version consistency so editing begins with verified context.
Helps connect Feishu or Lark bots, documents, spreadsheets, and approval workflows. Guides API authentication and MCP setup, a way to expose tools to agents, using environment variables for credentials and verification before externally visible changes.
Bring one approved community pull request into a release branch while preserving authorship and contribution credit. Review the changes, check for conflicts, run focused tests, and post a specific thank-you to the contributor.
Update a clearly specified set of dependencies by checking current versions and release notes, applying changes, fixing compatibility problems, and running the project's tests and build checks to verify the result.
Write or update repository and product documentation, including README files, user guides, API documentation, architecture notes, and migration instructions. Identify the audience, improve existing documents where possible, and verify commands and paths while flagging unknown details.
Guides users through removing a stored fact with Codewhale’s Context Lens memory controls. It identifies the intended fact, explains the forget action, and distinguishes confirmed deletion from an action that has not yet completed.
Find personal photos in a local macOS Photos library by album or date, then export selected copies to a chosen folder. Report exported and missing items while preserving originals and existing destination files.
Find reference images or photos for documents, pages, and presentations, then save selected files locally. Check that each file opens and report its location, dimensions, source, and stated license so the results are ready to review.
Search and read Gmail messages, identify items that need attention, and prepare email drafts. Send messages with explicit approval of recipients, subject, and body, using an authorized Gmail connection.
Provides a compatibility name that directs agents from presentations to the pptx skill. It identifies which skill to load for the actual workflow, helping users understand an existing reference to presentations.
Verify Codewhale changes with checks matched to their scope and risk. Collect actual test counts, automated limit checks, and command output to show what passed, what failed, and what remains unverified.
Is this your account? Link GitHub to bring it onto your creator page (Japanese)