coordinate-external-ag…
Coordinate independently operated external agents through durable handoffs. Use when work crosses hosts, sessions, accounts, services, queues, boards, pull…
Prepare, stage, and safely submit high-signal Apple Feedback Assistant bug reports, crash reports, enhancement requests, and Foundation Models feedback attachments. Use when Codex needs to gather reproducible evidence for Apple software, SDKs, developer tools, hardware,
$ npx -y skills add gaelic-ghost/socket --skill feedback-assistant-workflow --agent claude-codeHow it fires
How this skill gets triggered: by you, by Claude, or both.
/feedback-assistant-workflowContext preview
The summary Claude sees to decide when to auto-load this skill.
Prepare, stage, and safely submit high-signal Apple Feedback Assistant bug reports, crash reports, enhancement requests, and Foundation Models feedback attachments. Use when Codex needs to gather reproducible evidence for Apple software, SDKs, developer tools, hardware,
name: feedback-assistant-workflow description: Prepare, stage, and safely submit high-signal Apple Feedback Assistant bug reports, crash reports, enhancement requests, and Foundation Models feedback attachments. Use when Codex needs to gather reproducible evidence for Apple software, SDKs, developer tools, hardware, documentation, beta releases, or system-model behavior; operate the Feedback Assistant app or website; or turn a Foundation Models session into a reviewable feedback attachment.
Prepare one actionable Apple report at a time. Own report quality, evidence selection, privacy review, and the Feedback Assistant handoff; do not replace the specialist workflow that diagnoses the underlying problem.
1. Classify the report as a bug, crash, regression, enhancement, documentation issue, or Foundation Models behavior issue. Route security and privacy vulnerabilities to Apple's security reporting program instead of ordinary Feedback Assistant. 2. State one concise title, environment, minimal reproduction, expected result, actual result, frequency, and regression range when known. Keep unrelated issues in separate reports. 3. Build an attachment manifest before opening the submit path. Include source, collection time, purpose, and privacy review for each log, sample project, System Information Report, recording, screenshot, or `.json` feedback attachment. 4. Hand off diagnosis to the owning workflow when needed: `ios-runtime-forensics-workflow` for runtime evidence, `xcode-debugger-mcp-workflow` for a debugging session, `xcode-device-hub-workflow` for connected-device evidence, and `explore-apple-swift-docs` for API routing.
1. Prefer the Feedback Assistant app on Mac, iPhone, or iPad when timely diagnostics matter. Use the website only when the evidence was captured manually or the app is unavailable. 2. On this Mac, inspect `com.apple.appleseed.FeedbackAssistant`. Its signed-in main window exposes Inbox, Drafts, Submitted, Requests, team mailboxes when available, search, and New Feedback. 3. Select the narrowest topic and product area. Verify the report describes one issue before drafting or staging it. 4. Keep a staged report in Drafts until its prose and attachment manifest have been reviewed. Record the Feedback Assistant ID only after a report has been submitted. 5. If Apple requests more information, preserve the original report's scope and attach only the requested, reviewed evidence.
Use `LanguageModelSession.logFeedbackAttachment` only for Foundation Models behavior. It serializes a structured JSON attachment containing session transcript material and feedback metadata; it does not submit a report. Read `references/foundation-models-feedback-attachments.md`, review the resulting content, save it deliberately, and attach it through Feedback Assistant only after the privacy and submission gates above.
Stuff for Agents on macOS Promo audio: Socket Codex Marketplace Promo
Coordinate independently operated external agents through durable handoffs. Use when work crosses hosts, sessions, accounts, services, queues, boards, pull…
Assign worktree, branch, write, validation, integration, and cleanup ownership before parallel repository work. Use when a worker will inspect or modify…
Design framework-neutral agent and automation workflows before implementation. Use when choosing between Codex app automations, codex exec, Codex subagents,…
Design evaluation workflows for agent, skill, prompt, and automation behavior before implementation. Use when choosing eval cases, graders, thresholds,…
Design safe n8n workflows with deterministic routing, credentials, idempotency, recovery, local-model checks, drafts, and exact approval gates.
Coordinate bounded worker tasks with a launch envelope, report-back, escalation, and synthesis contract. Use before spawning, resuming, steering, cancelling,…