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 LFTPadilla/agent-dev-kit --skill explain-conceptgit clone --depth 1 https://github.com/LFTPadilla/agent-dev-kitWrote 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/lftpadilla/agent-dev-kit/explain-concept)<a href="https://agentmods.dev/skills/lftpadilla/agent-dev-kit/explain-concept"><img src="https://agentmods.dev/badge/skills/lftpadilla/agent-dev-kit/explain-concept/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/lftpadilla/agent-dev-kit/explain-concept"><img src="https://agentmods.dev/badge/skills/lftpadilla/agent-dev-kit/explain-concept.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 | $0.00107 | $0.00689 |
| Opus 5.5 | $0.00043 | $0.00276 |
| Sonnet 5.5 | $0.00021 | $0.00138 |
| Haiku 4.5 | $0.00011 | $0.00069 |
Grade A, and why
explain-concept 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 1mo 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 — 63 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Explain Concept 🧭
Umbrella router. It does not render anything. It picks the medium, hands off to one skill, and records what was learned.
When to use
- A request names an audience but not a format ("help the team understand the retry logic").
- A ticket, a bug, or a design must be explained to somebody who did not write it.
- You are about to write three paragraphs of prose about a structure. Stop and route.
Do not use it when the user already named the medium. Call that skill directly.
Step 1 — Clarify in ONE question
Ask one question only, and only if the answer is not already obvious:
Who is this for, and how will they receive it (image, editable file, or video)?
If the request already names an audience and a medium, skip the question.
Step 2 — Route
| Signal in the request | Route to | Output |
|---|---|---|
| Quick topology, flow, or architecture picture; paste into chat or a PR | diagram-render |
PNG |
| The reader must edit it, restyle it, or export SVG/PDF; swimlanes, UML, ER | drawio-skill |
.drawio + export |
| Narrated walkthrough, async share, non-technical audience, onboarding | explainer-video |
videos/<slug>/renders/video.mp4 |
| No image is possible or wanted; the reader will read the code | text trail | ordered file:line list |
Rules for the routes:
- diagram-render — the picture is static and the reader only looks at it.
- drawio-skill — the picture has a second life: somebody changes it later.
- explainer-video — timing matters. The reader needs the order of events spoken, not inferred. Example: a webhook race bug where two handlers write the same row.
- text trail — the fallback. Give an ordered list of
file:linestops, one sentence per stop on what to notice, and 5 lines of summary at most. Do not paraphrase the code.
Pick one route. Do not produce two mediums for the same request unless the user asks for both.
Step 3 — Always record lessons
After the chosen skill finishes, append one line to that skill's
references/lessons.md if anything was learned:
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.
- 1mo ago First seen · 63 lines · 107 tokens per session scan A b056a519b7b5
explain-concept is a skill published in the GitHub repository LFTPadilla/agent-dev-kit (2 stars, last pushed 27d ago), licensed MIT. It adds 107 tokens to every session and 689 once invoked, about $0.0004 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 2026-09-12.
Other skills, from other repositories
mindos
MindOS: local knowledge assistant & shared KB. Keeps decisions, notes, SOPs, debugging lessons, research findings, preferences across sessions/agents. Core: save notes, search KB, organize files, run workflows, review, append CSV, hand off context, distill lessons. NOT for app source or paths outside KB. Triggers…
coding-standards
Applies language-neutral engineering standards when writing or reviewing code, especially around boundaries, failures, domain models, dependencies, tests, and type safety.
flashcards
Turn notes or a topic into a set of question/answer study flashcards. Use when the user wants to revise or memorise material.
code-review
Review a code snippet for bugs, security issues, and style, with concrete fixes. Use when the user shares code and asks for feedback or a review.
explain-simply
Explain a complex topic in plain language with an everyday analogy. Use when the user wants something demystified or "explained simply".
pr-learning
Extract actionable learnings from merged PRs by comparing initial submission vs final merged state. Analyzes review cycles to capture what changed and why. Use when the user says "learn from PR", "extract PR learnings", "what did we learn from this PR", or wants to build a knowledge base from merged PRs.