agentleFS
matthewdonsemail-labmatthewdonsemail-lab/log/.claude/skills/log-discord-server-search/SKILL.md

log-discord-server-search

Skill63 starsChanged 2 days agoMITView on GitHub

Finding a conversation in a Discord server. A log play (docs/useCases/discord/serverSearch.md): Find what someone said in a server, or a conversation the owner had there, with Discord's own search across every channel, forum post and thread at once, then read around it. This is how \"the conversation I had with <someone> about <a topic> in the <server> server\" is found in one search instead of reading a channel back hundreds of messages. This page reads and sends nothing.

The file

We split this skill into the rules it gives an agent

2 rules. About 0 are common, 1 is distinctive, and 1 is 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.

  • 1

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    Search matches words: Use the words they would have typed: "widgets", not "generative UI components".

    line 36
  2. Very specific:

    The answer names where each hit is: Read it there with discordRead, which names the channel or post it read.

    line 38

≈ “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.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

Finding a conversation in a Discord server

This is the log play discord/serverSearch. Before anything else, call the log MCP tool getGuide({ useCase: "discord/serverSearch" }): it returns this page and makes it the session's play, so every log tool's answer says which step you are on and a send with no play loaded is refused (rules 157, 158). Then follow the steps in order.

Draft every message from that person's own thread and matt's real replies (voice with no text, search), never from memory, and never send two that read alike (sameAsSent).

Find what someone said in a server, or a conversation the owner had there, with Discord's own search across every channel, forum post and thread at once, then read around it. This is how "the conversation I had with <someone> about <a topic> in the <server> server" is found in one search instead of reading a channel back hundreds of messages. This page reads and sends nothing.

How the owner asks for it: "find the conversation I had with <someone> in <server> about <thing>", "what did <someone> say in <server> about <thing>", "search <server> for <words>", "read through the <server> server and find <thing>"

Run it from a fresh context

Say: "go on docs/useCases/discord/serverSearch.md for <server>, looking for <what>". The agent reads this page and runs the steps.

The steps

#StepToolDone when
1Search the server for the subject, narrowed to the person when the owner names onediscordSearch { account, server: "<server>", query: "<words>", from: "<who>" }hits with their time, author, text and where each one is ("<server> #<channel>" or a post)
2When nothing comes back, search again with other words the owner or they would have useddiscordSearch { account, server: "<server>", query: "<other words>" }the hit, or two searches with nothing
3Read the conversation around the hit, text only, in the channel or post it namesdiscordRead { account, channel: "<where the hit is>", limit: 60, noMedia: true }the whole exchange: what was asked and every answer
4Tell the owner what was saidthe reply to the ownerwho said what, when (the owner's time), quoted closely, and what it means for what the owner asked

What to know

  • Search first, always. Discord's search covers the whole server, including forum posts and threads; Log's own search only holds the mentions and replies the watcher kept from servers. Reading a channel back to look for something is the slow way.
  • Search matches words. Use the words they would have typed: "widgets", not "generative UI components". Try a second set before saying it isn't there.
  • from narrows to one person by name. It is matched on the hits' authors, so a short name works ("paul").
  • The answer names where each hit is. Read it there with discordRead, which names the channel or post it read.
  • Voice notes and images: a hit in a voice note shows as its attachment. Read that spot without noMedia to have it transcribed.
  • What each part of a server is: the discordServers skill in the discord guide.

Download now (ZIP)

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
What did you use it for? e.g. Reviewing pull requests in a Rails app

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.