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 Freespirits/social-auto-engine --skill socialblast-pipelinegit clone --depth 1 https://github.com/Freespirits/social-auto-engineWrote 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/freespirits/social-auto-engine/socialblast-pipeline)<a href="https://agentmods.dev/skills/freespirits/social-auto-engine/socialblast-pipeline"><img src="https://agentmods.dev/badge/skills/freespirits/social-auto-engine/socialblast-pipeline/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/freespirits/social-auto-engine/socialblast-pipeline"><img src="https://agentmods.dev/badge/skills/freespirits/social-auto-engine/socialblast-pipeline.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 | $0.00098 | $0.01192 |
| Opus 5 | $0.00049 | $0.00596 |
| Sonnet 5 | $0.00020 | $0.00238 |
| Haiku 4.5 | $0.00010 | $0.00119 |
Grade A, and why
socialblast-pipeline 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 12d 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 — 127 lines — stays where its author put it; the contents beside it link to each section on GitHub.
SocialBlast Pipeline
What this skill does
Given a one-sentence business description, you produce a 7-day social media campaign with AI captions, optional AI images, optional AI videos, and an optional virality score. Every output goes into the approval queue — no post ever publishes without the human pressing Approve. This is not optional. It is the architectural spine.
Tool surface
You have these MCP tools from server.py:
| Tool | Purpose |
|---|---|
socialblast_status |
Read which backends and platforms are configured |
socialblast_generate_campaign(business_description, platforms) |
Create 7 pending posts per platform from a single sentence |
socialblast_enrich_post(post_id, with_video=False) |
Run caption -> image (+ optional video) on one pending post |
socialblast_enrich_campaign(group_id, with_video=False) |
Batch enrich every pending post in a campaign |
socialblast_predict_virality(prompt, platform) |
Score a caption (requires HiggsField) |
socialblast_list_pending |
Read-only queue inspector |
The legacy post_to_facebook, get_page_posts, and 30+ other Facebook tools
are available too, but you should not call publish-style tools directly. The
approval queue is the only path to live.
Workflow
Step 1 — Check what's connected
Always start with socialblast_status. Tell the user what they have and what
they are missing. If video.active_backend is none and images.replicate
is also false, only captions will work. Be honest about that.
Step 2 — Get the business sentence
Ask one question:
What does your business do? One sentence is enough. For example: "Coffee shop in London that roasts in-house."
Wait for the answer. Do not generate placeholder content if the user has not answered.
Step 3 — Confirm platforms
Default to all five broadcast platforms: facebook, instagram, threads, linkedin, tiktok. Show the user the list and let them remove platforms they do not want. Skip WhatsApp unless they ask (it is direct-message, not broadcast).
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.
- 12d ago First seen · 127 lines · 98 tokens per session scan A c14d9a324667
socialblast-pipeline is a skill published in the GitHub repository Freespirits/social-auto-engine (25 stars, last pushed 1mo ago), licensed MIT. It adds 98 tokens to every session and 1,192 once invoked, about $0.0005 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 2026-08-30.
Other skills, from other repositories
mcp-host-styling-integration
Integrates MCP App UI with host theming system. Applies host CSS variables, handles onhostcontextchanged, safe area insets, display mode detection, and fullscreen configuration.
cog-knowledge-consolidation
Build structured knowledge frameworks from scattered vault notes with source attribution.
context-engineering
Dynamic context injection, mode switching (dev/review/research), selective loading, and strategic compaction for token optimization.
plan-implementation
Disciplined execution of approved plans with step-by-step verification, phase checkpoints, failure investigation, and mandatory code/security reviews.
consensus-mechanisms
Multi-protocol consensus for agent swarms supporting Raft leader election, Byzantine fault tolerance, Gossip state propagation, and CRDT conflict-free merging.
llm-classifier
LLM-based zero-shot and few-shot classification for flexible intent detection.