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 krishddd/social_dive --skill skillgit clone --depth 1 https://github.com/krishddd/social_diveWrote 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/krishddd/social_dive/skill)<a href="https://agentmods.dev/skills/krishddd/social_dive/skill"><img src="https://agentmods.dev/badge/skills/krishddd/social_dive/skill/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/krishddd/social_dive/skill"><img src="https://agentmods.dev/badge/skills/krishddd/social_dive/skill.svg" alt="Reviewed on agentmods" width="80" height="20"></a>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 | $0.00124 | $0.01118 |
| Opus 5.5 | $0.00050 | $0.00447 |
| Sonnet 5.5 | $0.00025 | $0.00224 |
| Haiku 4.5 | $0.00012 | $0.00112 |
Grade A, and why
social-dive 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 2mo 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 — 105 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Social Dive Skill
When to use this skill
Use Social Dive when the user asks you to:
- Research a topic across academic papers, code, or web sources
- Read content from a URL (paper, repo, article, video transcript, etc.)
- Search for papers, code, articles, or Q&A on a topic
- Read social/platform content (Twitter/X, Reddit, Bilibili, Xiaohongshu, Instagram, LinkedIn, Facebook, Xueqiu, Xiaoyuzhou, V2EX)
- Summarize content from any supported source
- Check which knowledge sources are currently available
Social channels are login-gated: doctor shows whether a backend (OpenCLI or a
platform CLI/cookies) is set up. If a social read returns error_code: unauthenticated, tell the user to run social-dive doctor and
social-dive configure --from-browser chrome — and to use a throwaway account.
How to use
Step 1: Check available channels (always first)
social-dive doctor --json
Returns each channel's status (ok/warn/off/error), its active_backend, and a
message. Prefer ok channels; a warn channel usually still works but is
degraded (e.g. missing API key). Backends change availability, so check first.
Step 2: Read one or more URLs
social-dive read <url> --format=json
social-dive read <url1> <url2> <url3> # fetched concurrently
social-dive read <url> --summarize
- A single URL is dispatched to the best channel for it and returns one JSON object; multiple URLs return a JSON array (fetched in parallel).
- Results include
url,title,backend, andfetched_at. On failure you get a structurederror_code(rate_limited,unauthenticated,restricted,timeout,not_found,error) instead of a crash — check it before using.
Step 3: Search across sources
social-dive search "transformer architecture" --channels=arxiv,semantic_scholar --limit=10 --format=json
social-dive search "Python async patterns" --channels=github,stack_overflow,devto
social-dive search "CRISPR gene therapy" --channels=pubmed,openalex,europe_pmc
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.
- 2mo ago First seen · 105 lines · 124 tokens per session scan A baab57399295
social-dive is a skill published in the GitHub repository krishddd/social_dive (1 stars, last pushed 4mo ago), licensed MIT. It adds 124 tokens to every session and 1,118 once invoked, about $0.0005 per session on Opus 5.5. 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 2026-08-31.
social-dive compared with skills from other repositories
| Mod | Type | Stars | Last push | Licence | Scan | Tokens | Works with |
|---|---|---|---|---|---|---|---|
| social-divekrishddd/social_dive | Skill | 1 | 4mo ago | MIT | A | 124 | Claude Code, Codex |
| bilibili_searchwhiteguo233/OpenBiliClaw | Skill | 3.7k | today | MIT | A | 20 | Claude Code |
| ai-seo-v2diegosouzapw/awesome-omni-skills | Skill | 171 | 4mo ago | MIT | A | 81 | Claude Code, Codex |
| web_researchDjtony707/TITAN | Skill | 19 | 4d ago | MIT | A | 13 | Claude Code, Codex |
| deep-researchSECRET4422/mcp-deep-research-server | Skill | 1 | 25d ago | MIT | A | 64 | Claude Code, Codex |
| Web Searchagentic-in/elephant-agent | Skill | 584 | 2mo ago | none | A | 21 | Claude Code, Codex |
| Workspace Searchagentic-in/elephant-agent | Skill | 584 | 2mo ago | none | A | 19 | Claude Code, Codex |