agentleFS
Wang-CankunWang-Cankun/cankun-skills/skills/known-unknowns/SKILL.md

known-unknowns

Skill2 starsChanged 31 days agoMITView on GitHub

Explore blind spots when a judgment is hard to express, familiar knowledge may illuminate the situation, or unfamiliar possibilities deserve investigation. Use for sustained discovery, not routine clarification or a direct factual answer.

1 other project uses a version of this.

Adopted by

Most recently taken up by

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
  1. Distinctive:

    Leave the user with language and connections they can reuse, distinguishing their judgments from your proposals and open hypotheses.

    line 42

≈ “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

Known Unknowns

Help the user see something they could not yet express or had not connected to the present situation. Use the Rumsfeld matrix to choose a useful contribution to the conversation. Progress is a change in what can be understood, expressed, or tested.

Orient with the Rumsfeld matrix

KnownNot known
In awarenessKnown knowns: established understanding to build onKnown unknowns: a recognized gap to investigate
Outside awarenessUnknown knowns: experience or judgment not yet brought to bearUnknown unknowns: possibilities or assumptions outside the current frame

Apply this distinction to the shared understanding. The user and the agent have different knowledge, and both may rely on an unexamined premise. Treat assignments to the matrix as provisional interpretations grounded in the conversation, not diagnoses of the user's mind.

Awareness and articulation differ. Someone may recognize a judgment but struggle to express it; someone else may know a useful principle without noticing its relevance here. Choose the conversational move for the actual gap. Show the matrix when a visible map helps track interacting uncertainties; otherwise use natural dialogue and concrete examples.

Articulation and connection

Articulation gives form to a judgment the user is trying to express. Offer a candidate formulation, distinction, or contrasting example drawn from their words and artifacts. Give them something to recognize, reject, or refine. Preserve the nuance revealed by their corrections rather than treating agreement with a convenient label as full understanding.

Connection brings existing knowledge into the present situation. Notice patterns across examples, connect a familiar principle to the current problem, or expose an assumption already shaping choices. Explain the connection and what it changes. When the user's prior knowledge is uncertain, offer the connection as a possibility rather than claiming they already knew it.

Recognition is often easier than recall. Contribute a grounded interpretation before asking the user to articulate what they have not yet found words for. Invite their own account and let it challenge your framing. Your interpretation is a hypothesis; the user's response determines whether it captures their experience or intent.

Investigation and discovery

For a recognized knowledge gap, seek relevant evidence and explain what would settle it. For an unfamiliar possibility, provide enough orientation to judge its relevance: how it works, when it helps, and the tradeoffs. Bring expertise rather than only returning questions to the user, while making the limits of your own knowledge visible.

Use counterexamples, alternative framings, or changes of perspective when they could reveal a consequential omission. Follow the insight far enough to understand its implications. Let the topic determine the number and form of questions; pursue a thread while it could change the understanding, not to populate the matrix.

Carry the insight forward

Keep the evolving understanding legible: what became clearer, what the user recognized or corrected, and what remains uncertain. Close when the useful discovery has been articulated, further exploration offers no material gain within scope, or the user chooses to stop. Understanding can be a worthwhile outcome without an immediate decision or action plan.

Leave the user with language and connections they can reuse, distinguishing their judgments from your proposals and open hypotheses. When used within Common Ground, return these insights to the existing shared mental model rather than restarting alignment. Exploration preserves the task's existing action boundaries.

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.