Ai commonsense reasoning
APM-installable agent skills for Lean 4 and Mathlib4 — proof tactics, math domains, review and research workflows, generic tooling.
npx -y skills add r-irbe/proof-skills --skill ai-commonsense-reasoningAssembled from the repository path, not quoted from the project. Check it against their README if it does not work.
One thing to look at
- 2 stars2 stars. Stars are a popularity signal and not a quality one, but at this level it is likely that nobody has read this closely except its author, and you would be relying on your own review.
What its author says it does
Copied from the file, not written here
USE FOR: Commonsense reasoning for AI systems — world knowledge, naive physics, folk psychology, temporal/spatial reasoning, default reasoning, and their formalization. Use for reasoning about everyday knowledge that humans take for granted but AI systems need explicitly, especially when a downstream project must make implicit context explicit. DO NOT USE FOR: formal KR/ontology engineering (use @ai-symbolic-neuro); causal/deontic reasoning (use @ai-causal-deontic); formalisation in Lean (use @lean-knowledge-formalization). TRIGGERS: commonsense, naive physics, folk psychology, temporal reasoning, spatial reasoning, default reasoning.
SKILL.md
4.0 KB, as published. Nobody here has run it
Commonsense Reasoning
Formalizing the vast body of everyday knowledge that underlies human reasoning and is critical for AI systems operating in real-world contexts.
Routing
- USE FOR: Commonsense reasoning for AI systems — world knowledge, naive physics, folk psychology, temporal/spatial reasoning, default reasoning, and their formalization. Use for reasoning about everyday knowledge that humans take for granted but AI systems need explicitly, especially when a downstream project must make implicit context explicit.
- DO NOT USE FOR: formal KR/ontology engineering (use @ai-symbolic-neuro); causal/deontic reasoning (use @ai-causal-deontic); formalisation in Lean (use @lean-knowledge-formalization).
- TRIGGERS: commonsense, naive physics, folk psychology, temporal reasoning, spatial reasoning, default reasoning.
Workflow
- Classify the commonsense gap: naive physics, folk psychology, temporal, spatial, or default reasoning.
- Pick the matching paradigm from the body (qualitative reasoning, situation calculus, default logic, etc.) and the representative formalism.
- Apply the formalism to the concrete scenario; document where it does or doesn't yield the human-intuitive answer.
- Hand off: to
@ai-symbolic-neurofor ontology grounding, to@lean-knowledge-formalizationfor a Lean encoding, to@lean-zettelkasten.
Recovery & STOP
- STOP if the question is purely about KR/ontology — delegate to
@ai-symbolic-neuro. - STOP if the answer would require world-model data not in the body — escalate to
@research-council. - STOP if the question is about causal/deontic norms — delegate to
@ai-causal-deontic.
Handoffs
- Predecessors:
agent:gateway,skill:lean-research. - Successors:
skill:ai-symbolic-neuro,skill:lean-knowledge-formalization,skill:lean-zettelkasten.
Detailed reference
Full content for ai-commonsense-reasoning lives in
references/ai-commonsense-reasoning-handbook.md.
Load that file when the skill is convened; the SKILL.md only carries
the dispatch contract and the parts index.
| Section | Topic |
|---|---|
| Part 1 | Commonsense Knowledge Domains |
| Part 2 | Formal Commonsense Frameworks |
| Part 3 | Qualitative Reasoning |
| Part 4 | Theory of Mind & BDI |
| Part 5 | Temporal Commonsense |
| Part 6 | Commonsense Physics & Spatial Reasoning |
| Part 7 | Integration with Lean Formalization |
| Part 8 | Research Directions |
| Part 9 | Skill Cross-References |
See also
../../references/ai-commonsense-reasoning-handbook.md— Full handbook (extracted from this skill)../ai-symbolic-neuro/SKILL.md— Successor../lean-knowledge-formalization/SKILL.md— Successor../lean-zettelkasten/SKILL.md— Successor