Getting it into your agent
It runs from inside its repository, so the clone comes first — what it calls does not travel with the file alone.
git clone --depth 1 https://github.com/xg-gh-25/SwarmAInpx agentmods add skills/xg-gh-25/swarmai/s_learn-contentWrote 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/xg-gh-25/swarmai/s_learn-content)<a href="https://agentmods.dev/skills/xg-gh-25/swarmai/s_learn-content"><img src="https://agentmods.dev/badge/skills/xg-gh-25/swarmai/s_learn-content/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/xg-gh-25/swarmai/s_learn-content"><img src="https://agentmods.dev/badge/skills/xg-gh-25/swarmai/s_learn-content.svg" alt="Reviewed on agentmods" width="80" height="20"></a>- NVIDIA SkillSpector warn
SkillSpector: 2 findings, up to medium
These are SkillSpector’s own severities. On a checked sample its high-severity flags on skills were ~96% false positives — a documented command, a public API, a “never do X” rule — so we show them as a caution to read, not a verdict. Why →
- medium Data Exfiltration · line 78 Data is being sent to an external URL. This could be legitimate telemetry or data exfiltration. Manual review is recommended.Fix: Verify the destination URL is trusted and necessary. Remove or replace with documented APIs. Ensure no secrets, tokens, or PII are transmitted.
- medium Data Exfiltration · line 82 Data is being sent to an external URL. This could be legitimate telemetry or data exfiltration. Manual review is recommended.Fix: Verify the destination URL is trusted and necessary. Remove or replace with documented APIs. Ensure no secrets, tokens, or PII are transmitted.
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 | $0.00074 | $0.05106 |
| Opus 5 | $0.00037 | $0.02553 |
| Sonnet 5 | $0.00015 | $0.01021 |
| Haiku 4.5 | $0.00007 | $0.00511 |
Grade A, and why
learn-content scanned grade A with 1 finding 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 today.
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.
Makes network callslowCapability
Not a fault in itself. Listed so you know the mod talks to something, and to what.
**Tier 2: curl with platform-specific UA** (works for ~20% more) How it starts
The opening of the file, as written. The whole thing — 417 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Learn Content
Ingest content from any source into a structured knowledge card — a lightweight index entry that stores the source, key insights, and tags. NOT full-text archival. The source URL is the single source of truth; the card is an index pointer with enough context to decide whether to re-read.
Storage
~/.swarm-ai/SwarmWS/Knowledge/Learned/YYYY-MM-DD-<slug>.md
One file per ingested item. No hard size limit — prioritize learning value over compactness.
Workflow
Step 1: Accept Input
User provides one or more of:
- URL (article, blog post, tweet thread, GitHub repo, WeChat article)
- Pasted text block
- File path (PDF, doc, etc.)
- Forwarded message with link
Detect the input type:
| Input | Action |
|---|---|
| URL (general) | 3-tier fetch chain (see below) |
| Text block (no URL) | Use directly — store as source_type: text |
| File path (video/audio) | ffmpeg extract audio → whisper-transcribe → text |
| File path (document) | Read tool / appropriate skill (s_pdf, s_docx) |
| Multiple URLs | Process each separately, one card per URL |
3-Tier Fetch Chain (BLOCKING — exhaust all tiers before asking user)
Every URL goes through this chain. Stop at the first tier that returns usable content.
Tier 1: WebFetch (fastest, works for ~70% of URLs)
- Standard fetch. If it returns real content, done.
- Skip to Tier 2 if: anti-scraping block, "环境异常", empty body, login wall, SPA shell (
<div id="app"></div>)
Tier 2: curl with platform-specific UA (works for ~20% more)
# WeChat articles (mp.weixin.qq.com)
curl -sL -H "User-Agent: Mozilla/5.0 (iPhone; CPU iPhone OS 17_0 like Mac OS X) AppleWebKit/605.1.15 (KHTML, like Gecko) Mobile/15E148 MicroMessenger/8.0.43" "<URL>"
# General anti-scraping (Douyin pages, news sites)
curl -sL -A "Mozilla/5.0 (Macintosh; Intel Mac OS X 10_15_7) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/131.0 Safari/537.36" -H "Accept-Language: zh-CN,zh;q=0.9" "<URL>"
- WeChat: extract from
og:tags +id="js_content"div - General: extract from
<article>,<main>, or largest text block - Skip to Tier 3 if: SPA with no server-rendered content, JS-only rendering, video page
What ships with it
1 file beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
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.
- today Changed 1039e130d083
- 12d ago First seen · 417 lines · 74 tokens per session scan A b91f88addb00
learn-content is a skill published in the GitHub repository xg-gh-25/SwarmAI (44 stars, last pushed today), licensed MIT. It adds 74 tokens to every session and 5,106 once invoked, about $0.0004 per session on Opus 5. A static security scan graded it A with 1 finding (makes network calls). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.
Other skills, from other repositories
agent-challenges
Agent skill for challenges - invoke with $agent-challenges.
tutorial
Interactive tutorial teaching Ouroboros hands-on.
animation
Author animated technical explainer diagrams as .anim.json files for Nimbalyst's Animation editor. Use when the user wants to animate a diagram, show how a system/protocol/algorithm behaves over time, build a motion explainer, or turn a static architecture diagram into something that plays.
subscription-setup
A setup workflow for creating recurring medical-learning subscriptions after a doctor’s profile has been registered.
exam-material-recommendation
A study-material recommendation tool for doctors preparing for professional exams or job-related training. It uses the person’s target, timeline, available study time, and specialty, while treating the official exam syllabus as the main reference.
nw-document
Creates evidence-based documentation following DIVIO/Diataxis principles. Use when writing tutorials, how-to guides, reference docs, or explanations.