agentleFS
Sign inStart free
calebnewtonusccalebnewtonusc/Chewbacca/skills/what-can-this-do/SKILL.md

what-can-this-do

Skill8 starsChanged 3 months agoMITView on GitHub

Propose things this kit could do here that the user has not thought to ask for. Use when the user asks what you can do, what they should automate, what they are missing, or says they do not know where to start. Also use on arriving in an unfamiliar repo, and when the user is stuck choosing what to work on.

The file

We split this skill into the rules it gives an agent

2 rules. About 0 are common, 2 are distinctive, and 0 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈2

    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
  1. Distinctive:

    Every other skill assumes the user already knows the task.

    line 8
  2. Distinctive:

    Never answer this from the skill list alone.

    line 14

≈ “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.

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

what-can-this-do

Every other skill assumes the user already knows the task. This one is for when they do not, which is most of the time, because nobody can hold 39 skills and a second brain in their head.

Read before proposing

Never answer this from the skill list alone. Read the ground:

  • The repo: what it is, what is half-finished, what has no tests.
  • ~/.claude/skills/ and ~/Chewbacca/skills/: what is installed and what is quietly unused.
  • The second brain, especially NOW.md for what is actually live and what is broken right now.
  • superassistant recent 20: what he has actually been asking. Repeated questions are the strongest signal available, because a question asked three times is a procedure waiting to be written.
  • ~/coursework and the calendar when the week is the constraint.

Return three, not ten

Three concrete proposals. For each one:

  • What it would do, in a sentence, in terms of his actual work.
  • The evidence. What in the files or the history says he needs this. A proposal with no evidence is a guess dressed up.
  • The cost. Roughly how long, and what it touches.
  • Why it is not already done. Often the honest answer is that it is a bad idea, and saying so is worth more than a fourth suggestion.

Rank them. Lead with the one you would actually do.

Bias toward what repeats

The best candidate is something done by hand more than twice. The second best is something currently failing silently. Novelty is the worst reason to propose anything.

Say what not to build

Close with at least one thing the kit could do and should not, with the reason. A list that only ever grows is a list nobody trusts.

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.

Check your setup, free