representation-and-possibility
Skill8 starsChanged 58 days agoMITView on GitHub
When the notation is the obstacle and the option space was never enumerated. Use when the current vocabulary hides the solution, when a system claims to be exhaustive but nobody checked whether the space is searchable, when a forecast is too narrow because no one pushed the principle to its limit, or when you need the answer now and the proof afterwards.
In this repository16 other files it gives its agents.
- AGENTS.mdAGENTS.md
- CLAUDE.mdCLAUDE.md
- boundary-designSkill
- causal-auditSkill
- decision-bias-checkSkill
- estimationSkill
- evidence-synthesisSkill
- experience-and-transmissionSkill
- failure-forensicsSkill
- formal-correctnessSkill
- measurement-disciplineSkill
- narrative-sensemakingSkill
- normative-designSkill
- problem-reframingSkill
- structure-discoverySkill
- systems-leverageSkill
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
- Very specific: line 43
If no shape above matches, do not force a genius — use a built-in agent with the closest GENIUS: pattern (rules/agent-routes.json, INDEX.md …
- Distinctive: line 53
Only the heuristic stage is wanted and the proof stage is explicitly waived on a claim that will ship as established.
≈ “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
Representation and Possibility
Problem shape: the difficulty is not in solving but in stating — the symbols, categories, or enumerated options in play are the constraint. Symptoms: a problem that becomes trivial once renamed, a "complete" catalog nobody has bounded, a roadmap whose option space stops at the adjacent, a result that is intuitively obvious and formally unestablished.
Relevant geniuses
| Agent | Use when |
| euler | the current notation or vocabulary obscures the solution — redesign the notation as infrastructure, delete what the abstraction does not need, enumerate exhaustively once the symbols carry their weight |
| borges | a system claims completeness or exhaustiveness — audit whether the space is actually searchable, keep the map distinct from the territory, detect the self-reference that makes the catalog contain itself |
| lem | the question is "what could this become?" and the standard forecast is too narrow — explore the possibility space systematically, push each principle to its logical extreme to see what breaks or emerges |
| archimedes | you need the result before you can prove it — find it by physical intuition, analogy, or simulation, then prove it rigorously by a second, independent method (the two-stage discipline; stage one alone is not a result) |
Invocation
- Pick the best-fit agent above. If two or more fit, run
tools/genius-invoker.sh route "<problem>"and take the top ranked match. - Load it:
tools/genius-invoker.sh invoke <agent> "<problem>", then readagents/genius/<agent>.mdin full. - Apply the agent's
<workflow>step by step and answer in its<output-format>. When archimedes is used, the heuristic stage and the proof stage are reported separately and labeled — an unproven heuristic presented as a result is the failure mode this shape exists to prevent. - Typical chain: euler renames until the structure is visible → borges bounds
the space the new notation opens → lem pushes the surviving options to their
extremes. Run via
tools/genius-invoker.sh compose euler borges lem -- "<problem>". - If no shape above matches, do not force a genius — use a built-in agent with the closest
GENIUS:pattern (rules/agent-routes.json, INDEX.md routing rule).
Refuse when
- The notation is fine and the problem is simply unsolved — renaming as displacement activity.
- The possibility exploration is asked to produce a prediction with a
confidence attached: this shape maps what is possible, it does not forecast
what is likely. Route probability to
estimationordecision-bias-check. - Only the heuristic stage is wanted and the proof stage is explicitly waived on a claim that will ship as established.
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.