Agent-Skills.md

Agent Skills: Datalog Fixpoint Skill

Datalog bottom-up fixpoint iteration for recursive queries

UncategorizedID: plurigrid/asi/datalog-fixpoint

Repository

plurigrid/asi
plurigridLicense: MIT
165

Install this agent skill to your local

pnpm dlx add-skill https://github.com/plurigrid/asi/tree/HEAD/plugins/asi/skills/datalog-fixpoint

Skill Files

Browse the full folder contents for datalog-fixpoint.

Download Skill

Loading file tree…

plugins/asi/skills/datalog-fixpoint/SKILL.md

Skill Metadata

Name
datalog-fixpoint
Description
Datalog bottom-up fixpoint iteration for recursive queries

Datalog Fixpoint Skill

Bottom-up fixpoint iteration for recursive Datalog queries without explicit recursion.

Core Concept

Datalog computes fixpoints via iterative saturation:

T^0(∅) → T^1 → T^2 → ... → T^ω (fixpoint)

Where T is the immediate consequence operator.

Scientific Skill Interleaving

This skill connects to the K-Dense-AI/claude-scientific-skills ecosystem:

Dataframes

  • polars [○] via bicomodule
    • High-performance dataframes

Bibliography References

  • algorithms: 19 citations in bib.duckdb

Cat# Integration

Fixpoint computation maps to Cat# via coalgebraic semantics:

Trit: 0 (ERGODIC - iterative bridge)
Home: Prof (profunctors/bimodules)
Poly Op: ⊗ (parallel saturation)
Kan Role: Adj (Kleisli adjunction)

GF(3) Naturality

Datalog fixpoint iteration is inherently ERGODIC:

  • Each iteration step is a natural transformation
  • Convergence = reaching the terminal coalgebra
  • The fixpoint IS the bicomodule equilibrium