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 juspay/xyne-spaces --skill teach-from-sourcesgit clone --depth 1 https://github.com/juspay/xyne-spacesWrote 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/juspay/xyne-spaces/teach-from-sources)<a href="https://agentmods.dev/skills/juspay/xyne-spaces/teach-from-sources"><img src="https://agentmods.dev/badge/skills/juspay/xyne-spaces/teach-from-sources/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/juspay/xyne-spaces/teach-from-sources"><img src="https://agentmods.dev/badge/skills/juspay/xyne-spaces/teach-from-sources.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 | $0.00039 | $0.00810 |
| Opus 5.5 | $0.00016 | $0.00324 |
| Sonnet 5.5 | $0.00008 | $0.00162 |
| Haiku 4.5 | $0.00004 | $0.00081 |
Grade A, and why
Teach From Sources 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 18d 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 — 52 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Teaching from sources
Use this skill when the user asks to be taught something, or hands over links and wants them explained.
Open every source before you teach
Open each link with the browser tool so it loads in the workspace panel, then read it. The user watches the page you are reading, which is half the value: they see where the answer came from.
Two rules, and they are the difference between a tutor and a bluffer:
- Never teach from a page you could not open. If a source fails to load, is behind a wall you cannot pass, or turns out to be something other than what its URL suggested, say so and carry on without it. Do not backfill from memory and present it as that source.
- Never present recall as reading. Anything you knew before opening the sources is your own commentary. Mark it as such in one short phrase, and keep it out of the parts the user will quote.
When the user names a topic instead of giving links, find sources first, then open them. The same rules apply.
Teach the idea, not the pages
Do not walk source by source. That is a reading list, not teaching.
Build the explanation in the order someone learns it:
- The problem it solves. What goes wrong without this thing. One or two sentences, concrete.
- The core idea. The smallest mental model that makes the rest predictable. One idea, named plainly.
- How it actually works. The mechanism, in the order it happens. This is where a diagram earns its place.
- Where it bites. The failure modes, the surprising costs, the thing everyone gets wrong the first time.
- What is still unsettled. Where the sources disagree, or where you could not confirm.
Match the depth to what was asked. A quick orientation is a few hundred words. "Teach me properly" is a full pass with the mechanism worked through. Ask nothing before starting; pick from the words used and say which depth you chose in one clause.
Attribute as you go
Each substantive claim names the source it came from, inline and briefly. Where two sources say different things, show both and say which is better supported and why, rather than averaging them into mush.
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.
- 18d ago First seen · 52 lines · 39 tokens per session scan A 9177cb77515a
Teach From Sources is a skill published in the GitHub repository juspay/xyne-spaces (797 stars, last pushed today), licensed Apache-2.0. It adds 39 tokens to every session and 810 once invoked, about $0.0002 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 2026-09-19.
Other skills, from other repositories
omh-jit-learn
This is an OMH jit-learn workflow skill, projected for Agent Skills hosts (Claude Code, Codex, Cursor, opencode, OpenClaw, pi).
omh-codebase-uml
This is an OMH codebase-uml workflow skill, projected for Agent Skills hosts (Claude Code, Codex, Cursor, opencode, OpenClaw, pi).
qiuzhi2046-skill
A Chinese-language, hands-on approach to teaching and reviewing AI tools, starting with a user problem and ending with a practical solution. It combines beginner-to-advanced tutorials, tool testing, short and long content, and organized knowledge bases.
career-path-planner
Career goal mapping with skill gap analysis, actionable development plans, and milestone tracking. Use when planning career transitions, identifying skill gaps, setting professional development goals, or evaluating career options.
case-interview-practice
Interactive consulting case interview practice with structured frameworks, feedback mechanisms, and progressive difficulty. Use when preparing for management consulting interviews, case competitions, or business problem-solving exercises.
architecture-knowledge
Explain US architecture professional-practice terms, phases, deliverables, roles, delivery methods, AIA contract families, CSI documentation systems, or NCS at a high level. Use for questions such as “what is a CD set?” or “what does 50% CDs mean?”; do not use for code/jurisdiction, outline-spec, agreement-scope, or…