Add Xquik social data skill (#97)

Co-authored-by: kriptoburak <kriptoburak@users.noreply.github.com>

skills/xquik-social-data/SKILL.md

@@ -0,0 +1,82 @@
+---
+name: xquik-social-data
+description: "Use Xquik for X and Twitter social data workflows through its public API, SDKs, MCP server, webhooks, and installable agent skill."
+version: 0.1.0
+author: kriptoburak
+tags: [xquik, twitter, x, social-data, api, mcp]
+---
+
+# Xquik Social Data
+
+Use this skill when a user needs to collect, normalize, monitor, or automate X and Twitter data with Xquik. Xquik provides a public REST API, generated SDKs, an HTTP MCP server, webhooks, and an installable agent skill for common social data workflows.
+
+## When to Use
+
+- Search tweets, inspect tweet details, or collect account timelines
+- Fetch user profile data, followers, following, mentions, media, or engagement data
+- Run bulk extraction jobs for replies, quotes, posts, media, lists, communities, or people search
+- Set up monitors and webhook delivery for new social events
+- Use Xquik from an AI agent through the public MCP server or installable skill
+- Draft, schedule, or confirm write actions only when the user explicitly asks
+
+## Required Inputs
+
+- `XQUIK_API_KEY` for API, SDK, or MCP calls
+- The target endpoint, task type, username, tweet URL, tweet ID, query, or extraction type
+- The desired output shape, such as JSON, CSV, summary, dashboard table, or webhook payload
+- User confirmation before private reads, write actions, monitor creation, webhook delivery, or bulk jobs
+
+## Workflow
+
+1. Read the public Xquik docs before selecting endpoints. Start with the API reference for REST routes and the MCP guide for agent setup.
+2. Use the installable skill when the agent supports Skills:
+
+   ```bash
+   npx skills@1.5.3 add Xquik-dev/x-twitter-scraper
+   ```
+
+3. For JavaScript or TypeScript helpers, pin the validated package version:
+
+   ```bash
+   npm install x-developer@2.4.16
+   ```
+
+4. Keep credentials in environment variables or the host secret store. Never paste API keys into prompts, logs, source files, PRs, or issue text.
+5. Choose the narrowest endpoint or extraction type that satisfies the task. Do not fetch extra pages, private data, or write-capable resources without user approval.
+6. Preserve pagination metadata such as `next_cursor` and `has_more`. For long jobs, estimate first, start the job, then poll the documented job endpoint until it finishes or fails.
+7. Normalize outputs before analysis. Keep raw IDs, source URL, collected-at time, query parameters, and pagination state so results can be audited later.
+8. For monitors and webhooks, confirm the target account or keyword, event types, destination URL, and ongoing behavior before creating resources.
+
+## Output Shape
+
+Return structured results unless the user asks for prose only:
+
+```json
+{
+  "source": "xquik",
+  "task": "tweet-search",
+  "query": "from:example launch",
+  "items": [],
+  "has_more": false,
+  "next_cursor": null,
+  "notes": []
+}
+```
+
+For summaries, include the query, time window, result count, missing fields, and any follow-up cursor or job ID.
+
+## Safety Checks
+
+- Stop when the task needs an unsupported endpoint or unavailable data. Do not guess response fields.
+- Ask for explicit confirmation before write actions, private reads, monitor creation, webhook delivery, or large extraction jobs.
+- Do not log, echo, or commit `XQUIK_API_KEY`.
+- Do not present internal implementation details. Describe the public API, SDKs, MCP server, webhooks, and documented workflows only.
+- Check public links and package availability before adding install snippets to docs or generated output.
+
+## References
+
+- Xquik documentation: https://docs.xquik.com
+- API reference: https://docs.xquik.com/api-reference/overview
+- MCP guide: https://docs.xquik.com/mcp/overview
+- Source repository and installable skill: https://github.com/Xquik-dev/x-twitter-scraper
+- npm registry package: https://registry.npmjs.org/x-developer