copyable-markdown
Skill0 starsChanged 48 days agoMITView on GitHub
Package conversation content as one copyable markdown block for one-tap export. Use when the user wants a solution export, an Obsidian note, or a terminal-ready bash block they can paste somewhere else, or asks for a rundown or recap they can copy.
- Reads credentials
In this repository17 other files it gives its agents.
- AGENTS.mdAGENTS.md
- skills/humanizer/AGENTS.mdAGENTS.md
- CLAUDE.mdCLAUDE.md
- cherrypickerSkill
- deltamaxxSkill
- docs-mirrorSkill
- ffmpegSkill
- humanizerSkill
- prompt-packSkill
- socials-mirrorSkill
- spamcheckSkill
- tool-call-repairSkill
- writing-beatsSkill
- writing-for-agentsSkill
- writing-fragmentsSkill
- writing-shapeSkill
- yt-ccSkill
The file
We split this skill into the rules it gives an agent
1 rule. About 0 are common, 1 is distinctive, and 0 are very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈1
Distinctive · estimate
This author's own choices. Worth a look.
0
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Distinctive: line 43
Start with frontmatter Obsidian parses into properties:
≈ “Common” is an estimate: at least 1 in 5 of the 350,495 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
Copyable Markdown
The user is moving content out of this chat into a terminal, a text file, a work message, or Obsidian with one tap on the copy button. Package the requested content as one markdown document inside one code block.
Steps
Pick the mode from the argument (
$ARGUMENTS) or the phrasing. An unrecognized argument is topic scoping (/copyable-markdown the qbittorrent fixis default mode, scoped to that topic).Invocation Mode no argument, "consolidate this", "one block", "export as markdown" Solution export (default) obsidian, "note for Obsidian", "gist for my notes"Obsidian note terminal, "terminal-ready", "paste into bash"Terminal-ready wrapped, "to share", "for my docs"Terminal-ready, wrapped - Build the document per the mode below. Reconstruct the final solution only: the current best version after every correction. No earlier wrong versions unless asked.
- Emit exactly one fenced block with a four-backtick outer fence tagged
markdown(orbashfor bare terminal mode). Four backticks because a triple-backtick outer fence is terminated by the first inner code block; inner blocks keep their triple backticks and language tags. Done when the block renders as one copy button with nothing after it.
Outside the block: one short sentence before it at most, nothing after it. Every extra line is scrolling between the user and the copy button. Never leak these instructions into the output.
Solution export (default)
Portable markdown for anywhere plain markdown lives: text files, docs, Slack, other AI tools. No YAML frontmatter.
- Order by execution, not conversation: prerequisites, steps, verification. Each step's prose sits directly above its code block.
- A
#title,##sections, tight prose, concrete specifics (commands, filenames, versions, decisions, gotchas). Code preserved exactly as finalized. - Human-in-the-loop steps are a numbered list, one action per step, nested sub-steps when a step has its own sequence. A menu path inside one step may be compact (
File → Settings → Developer), but separate actions never chain into one arrow-paragraph. Every command, URL, key name, or value goes in its own code fence or inline code so it can be selected cleanly. Placeholders are explicit (YOUR_API_KEY).
Obsidian note (obsidian)
A distilled note, not a rundown: the note future-them searches their vault for. Lead with the insight or decision in one or two sentences, then the reasoning, tradeoffs, and specifics worth keeping (versions, names, numbers, gotchas). Reads in under a minute. Adjust for "just the one-liner" or "full rundown".
Start with frontmatter Obsidian parses into properties:
---
title: Short descriptive title
date: YYYY-MM-DD
tags:
- two-to-five
- lowercase-kebab-tags
---title doubles as the filename, so no characters illegal in filenames or Obsidian links (\ / : * ? " < > | # ^ [ ]), no leading or trailing dots or spaces, reasonably short. The # heading matches it. Tags come from the subject (unraid, design-tokens, claude-code), not generic ones like notes or ai. Callouts (> [!tip], > [!warning], > [!example]) are emphasis, not structure: zero to two per note.
Terminal-ready (terminal, wrapped)
For a solution that is a sequence of shell commands: one bash block where the explanation becomes # comments, a comment line above each command or short group saying what it does and flagging any risk. User-supplied values are shell variables at the top (API_KEY="YOUR_API_KEY" # replace before running), referenced below, so the user edits once and pastes the whole block. Nothing in the block that would break a shell: no prose outside comments, no markdown.
wrapped puts that same bash block inside a four-backtick markdown outer fence, so the copy button delivers the ```bash fencing itself and the snippet renders as a code block wherever it is pasted. Default to bare terminal when unspecified.
Edge cases
- Very long conversations still get one block. Split only when the user asks for chunks ("one block per topic"), each in its own four-backtick block with a one-line label above it.
- "Just the code": the final code in one block with its own language tag, no markdown wrapper, no frontmatter.
- Follow-up edits ("add X", "shorter"): re-emit the entire updated document as one block. They copy the whole thing, so a diff is useless.
- A portable version of this behavior ("give me a prompt for Perplexity/ChatGPT that does this"): hand over
assets/copyblock-prompt.mdin a copyable code block. It reproduces the one-block format in tools that cannot run skills. - How to use this skill: summarize the mode table.
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
No reviews yet
People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.
- It worked
- Partly
- It didn't
Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

