Agent skills for Claude Code

End-to-end feature development orchestration from planning through deployment with quality gates

Replace Fedora's default zram swap with zswap plus a disk swap file, so heavy memory pressure causes slowness instead of OOM-killed apps.

Interact with Feed (feed.market): A2A and MCP endpoints for prediction markets, perpetuals, social feed, messaging, portfolio, and more.

User correction, thumbs up/down, inline editing, and reinforcement signals.

Design confirmations, status updates, and notifications that tell users an action registered.

Give effective feedback that promotes learning and growth.

Sweep configured feedback sources (Slack, GitHub Issues; email experimental) for new items: acknowledge at source, analyze recordings…

Feedback synthesis — cluster support tickets, NPS verbatims, app store reviews, and churn surveys by theme, separate signal from noise…

Extract Feishu (Lark) Docs, Wiki pages/collections, spreadsheets, and Minutes (妙记) transcripts into faithful local Markdown via the…

Send notifications to Feishu/Lark. Internal utility used by other skills, or manually via /feishu-notify. Use when user says "send to Feishu

Manage Feishu Wiki documents by creating, updating, and organizing content directly through Claude Code commands.

Generate or edit images using OpenAI GPT Image API (gpt-image-2, gpt-image-1, etc).

Review code, PRs, diffs, or implementation plans for distributed-systems risks, HTTP API contracts, queue consumers, scheduled jobs, and…

OpenTelemetry Semantic Conventions expert.

Audit a Claude skill for quality, context-efficiency, and best-practice compliance — classify its archetype, score the four principles and…

Automate Felt tasks via Rube MCP (Composio).

Fetch a Prow job run summary from Sippy showing all failed tests grouped by SIG with error messages

Retrieve the latest Runway API reference from docs.dev.runwayml.com and use it as the authoritative source before any integration work

Reviews and authors Cloudflare Workers code against production best practices.

Fetch pull requests that are new in a given OpenShift payload compared to the previous payload

Fetch unresolved review comments from a PR and evaluate with receive-feedback skill