explain-message
Skill0 starsChanged 8 days agoMITView on GitHub
Explain a Slack message, thread, or DM that the user does not understand, and say what it asks of them. Use when the prompt is a Slack link plus "help me understand this", "I don't understand this message", "what is X asking", "what does this mean", "is <name> right?", "help me with this:", "how should I respond", or a pasted message from a colleague or their AI agent.
In this repository3 other files it gives its agents.
- procedure-writing-rulesSkill
- social-researchSkill
- tool-fitSkill
The file
We split this skill into the rules it gives an agent
3 rules. About 0 are common, 3 are distinctive, and 0 are very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈3
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 14
If the message names a project the user works on, find its folder or repo and read its README or notes file.
- Distinctive: line 15
Write the answer in the shape below.
- Distinctive: line 16
Check the answer against the gate below before you send it.
≈ “Common” is an estimate: at least 1 in 5 of the 364,980 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
explain-message
The user reads many messages written in someone else's shorthand (a salesperson's pipeline terms, an engineer's code terms, a colleague's AI agent output). The answer must land on the first read.
Procedure
- Read the full thread with your Slack tool (for example
slack_read_thread), not only the linked message. Read the 10 messages before it in the channel or DM if the thread starts mid-topic. If the user pasted the message and you have no Slack tool, work from the paste and say so. - Open every link, file, PR, or doc in the message that the meaning depends on. Read at most 3.
- If the message names a project the user works on, find its folder or repo and read its README or notes file.
- Write the answer in the shape below. Do not add other sections.
- Check the answer against the gate below before you send it.
Answer shape
- What they're asking you: one line. Say "nothing, FYI only" when that is the case.
- What the message says, in plain words: 2-4 sentences. Use no code identifiers, acronyms, or internal names that the user did not use. Define a term you must keep in the same sentence.
- One concrete example: one real entity (a customer, a lead, a page) that shows what the message means.
- Your call: the decision or reply it needs, with a recommendation.
- Draft reply: ready to paste. If it asks someone for something, open with their name and the action, then the deadline. Use only facts and dates the user gave.
Gate
Could the user act on the answer without asking "what does that mean"? If a sentence holds a term they did not use, rewrite the sentence.
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.

