Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx skills add alijancb/subio-mcp --skill social-scrapegit clone --depth 1 https://github.com/alijancb/subio-mcpWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/skills/alijancb/subio-mcp/social-scrape)<a href="https://agentmods.dev/skills/alijancb/subio-mcp/social-scrape"><img src="https://agentmods.dev/badge/skills/alijancb/subio-mcp/social-scrape/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/skills/alijancb/subio-mcp/social-scrape"><img src="https://agentmods.dev/badge/skills/alijancb/subio-mcp/social-scrape.svg" alt="Reviewed on agentmods" width="80" height="20"></a>- NVIDIA SkillSpector pass
What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00060 | $0.01185 |
| Opus 5 | $0.00030 | $0.00593 |
| Sonnet 5 | $0.00012 | $0.00237 |
| Haiku 4.5 | $0.00006 | $0.00119 |
Grade A, and why
social-scrape scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 7d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 96 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Social scrape
Collect what specific accounts and specific topics are actually saying, using the
subio-scrape MCP tools, and hand back either a chat summary or a document.
Every read happens in a local browser that is never signed in. Say so if the user asks whether it will touch their account: it does not, it clears session cookies before every run, and it does not solve challenges or disguise itself as a human. That is also the reason for the ceiling in step 2 — work within it rather than around it.
1. Find out what to read
If the user already named accounts, a topic or a hashtag, skip straight to step 2.
Otherwise ask with AskUserQuestion — one question, these options:
- Accounts — specific people or companies on X
- Topic or hashtag — a subject, wherever it is being discussed
- A company — one organisation across X and LinkedIn
- Open discussion — what Hacker News is saying
Then ask for the handles or the topic itself in a normal follow-up question. Also settle how far back (default: last 7 days) and how many items per source (default: 20) unless the user already said.
2. Pick the right tool for each kind
This is the part that decides whether the scan works, so read it before calling anything.
| The user wants | Call | Why this and not something else |
|---|---|---|
| Named X accounts | x_account_posts |
Signed out, X serves account timelines in full, with exact counts |
| A topic or hashtag on X | x_account_posts with match: ["term", "#tag"] over the accounts likely to discuss it |
X does not serve search, hashtags or explore to a signed-out browser. There is no search tool because there is no signed-out search. Reading the accounts and filtering is the way |
| A topic, hashtag or discussion generally | hackernews_search |
Hacker News publishes discussion openly through a real API — no browser, no session, no ceiling |
| A company | linkedin_page for the profile, x_account_posts for what it posts |
LinkedIn's public page carries the firmographics; X carries the posting |
| A company's posts | linkedin_company_posts |
Best effort — LinkedIn often withholds these signed out |
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 7d ago First seen · 96 lines · 60 tokens per session scan A b914b3fe2bec
social-scrape is a skill published in the GitHub repository alijancb/subio-mcp (35 stars, last pushed 10d ago), licensed MIT. It adds 60 tokens to every session and 1,185 once invoked, about $0.0003 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-09-04.
Other skills, from other repositories
plasmate
Browse the web using Plasmate for 16x fewer tokens than raw HTML. Use when fetching documentation, reading web pages, extracting structured data, or interacting with websites.
playwright-recording
Record browser interactions as video using Playwright. Use for capturing demo videos, app walkthroughs, and UI flows for Remotion videos. Triggers include recording a demo, capturing browser video, screen recording a website, or creating walkthrough footage.
dev-browser
Browser automation with persistent page state. Use when users ask to navigate websites, fill forms, take screenshots, extract web data, test web apps, or automate browser workflows. Trigger phrases include "go to [url]", "click on", "fill out the form", "take a screenshot", "scrape", "automate", "test the website"…
telnyx-meeting-bot
Use when an agent must join, observe, react in, transcribe, summarize, or follow up on a Zoom, Google Meet, Microsoft Teams, or Webex meeting with Telnyx Meeting Bot. Handles vague requests, request-specific live polling, name/phrase and semantic triggers, explicitly authorized speak/chat actions, recovery, and all…
Agent Browser Automation
Fast Rust-based headless browser automation CLI with Node.js fallback for AI agents, featuring navigation, clicking, typing, snapshots, and structured commands optimized for agent workflows.
Karma Testing
Comprehensive Karma test runner skill for browser-based JavaScript unit testing with Jasmine, Mocha, or QUnit frameworks, real browser execution, coverage reporting, and CI/CD pipeline integration.