l-space
L-Space Skill: The Library of All Libraries
DeepseekModel
Curated skill
Quality Excellent · 78
v1.0.0
Get
https://deepseekmodel.com/api/download.php?id=plurigrid-asi-skills-l-space-skill-md&format=skill
Download .skill
Standard format with system_prompt and model_config, ready for any agent framework
The actual content of the system_prompt field in the .skill file.
name l-space description L-Space Skill: The Library of All Libraries version 1.0.0 L-Space Skill: The Library of All Libraries Status : ✅ Production Ready Trit : 0 (ERGODIC - the Library itself is the coordinator) Color : #8B4513 (Leather brown - the color of infinite shelves) Principle : Books = Knowledge = Power = Energy = Mass Frame : All libraries connected across spacetime via narrativium coherence Overview L-Space (short for Library-Space) implements Terry Pratchett's theory of informational physics combined with Bumpus's categorical narrative theory. In L-Space: "Books bend space and time. One good bookshop can be a gate to a thousand worlds." The skill enables navigation through the infinite library that connects all repositories of knowledge across spacetime, while maintaining the three sacred rules: Silence in the Library Books must be returned by the date stamped Do not interfere with the nature of causality Pratchett's Informational Physics The Equation Books = Knowledge = Power = Energy = Mass ∴ A sufficient quantity of books distorts spacetime itself This is not metaphor. In sufficiently large libraries: Time flows differently between sections Spatial topology becomes non-Euclidean Bidirectional causation becomes possible Critical Mass Phenomena "A large enough collection of books creates its own gravitational well, drawing in more books, more knowledge, until it punches through into L-Space." When knowledge density exceeds threshold: The Octavo reads the reader - books become autonomous Sourcery - reality becomes substrate for narrativium Undelivered letters create fatal illusions class InformationMass CRITICAL_DENSITY = 1e6 # books per cubic meter def l_space_accessible? ( library ) book_density(library) >= CRITICAL_DENSITY end def distortion_factor ( library ) return 0 unless l_space_accessible?(library) Math .log(book_density(library) / CRITICAL_DENSITY ) end end Narrativium: The Story Force Pan Narrans "Humans are not Homo sapiens, the wise man. We are Pan narrans, the storytelling ape." Narrativium is the fundamental force that makes stories cohere. Every element contains its story - how it came to be, what it does, where it's going. class Narrativium # The coherence force that makes stories work def story_tension ( element ) # Every element resists violation of its narrative arc element.expected_trajectory - element.current_position end def narrative_collapse! ( elements ) # When stories lose coherence, reality stutters elements.each do | e | e.phase_space_position = e.expected_trajectory.terminal end end end Story Phase Space Stories map the phase space of existence: phase_space(story) = { beginning: initial_conditions, middle: trajectory_through_possibility, end: attractor_basin } Bumpus Categories of Narratives Sheaves on Posets of Intervals From Bumpus et al.: Narratives are sheaves on posets of intervals : @present SchNarrative(FreeSchema) begin Interval::Ob # Time windows Snapshot::Ob # State at instant Relationship::Ob # How snapshots relate source::Hom(Relationship, Snapshot) target::Hom(Relationship, Snapshot) timestamp::Hom(Snapshot, Interval) # Sheaf condition: snapshots agree on overlaps end Two Perspectives Perspective Sheaf Type Interpretation Cumulative Colimit-style "Everything that happened up to now" Persistent Limit-style "What persists across time" The Librarian navigates between these: def librarian_navigate ( from_book, to_book ) if cumulative?(from_book) && persistent?(to_book) # Must cross perspective boundary find_interval_isomorphism(from_book.intervals, to_book.intervals) end end Object-Agnostic Narratives Narratives work for any structure: Graphs (character interaction networks) Groups (symmetries preserved through story) Databases (consistency across commits) Repositories (version control as narrative) # The same sheaf machinery works for all narrative_of_graph(G::Graph) = StrDecomp(G) narrative_of_group(G::Group) = StrDecomp(cayley_graph(G)) narrative_of_repo(R::GitRepo) = StrDecomp(commit_dag(R)) The Librarian Protocol Orangutan Epistemology The Librarian (formerly a wizard, now Pongo pongo) guards L-Space. As a member of the Librarians of Time and Space: "The truth isn't easily pinned to a page. In the bathtub of history, the truth is harder to hold than the soap." module LibrarianProtocol ACCEPTABLE_RESPONSES = [ "Ook" , "Ook?" , "Ook!" , "Ook." ] def validate_query ( query ) # The Librarian knows. He doesn't explain. query.well_formed? && !query.violates_causality? end def traverse_l_space ( from: , to: , via: :triangle_inequality ) path = find_path_through_shelves(from, to) validate_no_temporal_paradox!(path) path end end Navigation Rules Never call the Librarian a monkey (he's an ape) Bananas are acceptable currency for difficult queries Some paths are one-way (temporal direction matters) The Library is bigger on the inside (always) Integration Architecture GF(3) Triads L-Space (0) participates in balanced triads: sheaf-cohomology (-1) ⊗ l-space (0) ⊗ glass-bead-game (+1) = 0 ✓ structured-decomp (-1) ⊗ l-space (0) ⊗ random-walk-fusion (+1) = 0 ✓ persistent-homology (-1) ⊗ l-space (0) ⊗ topos-generate (+1) = 0 ✓ Skill Integration Map ┌─────────────────────────────────────────────────────────────────┐ │ L-SPACE (ERGODIC 0) │ │ The Library Coordinates All │ ├─────────────────────────────────────────────────────────────────┤ │ │ │ ┌──────────────────┐ ┌──────────────────┐ │ │ │ sheaf-cohomology │◄────────────►│ structured-decomp│ │ │ │ (-1) │ LOCAL→GLOBAL│ (0) │ │ │ │ │ │ │ │ │ │ Čech cohomology │ │ Tree decomp │ │ │ │ for consistency │ │ FPT algorithms │ │ │ └────────┬─────────┘ └────────┬─────────┘ │ │ │ │ │ │ │ ┌───────────────────┐ │ │ │ └────►│ L-SPACE │◄─────┘ │ │ │ │ │ │ ┌──────────────┤ Narrativium glue ├──────────────┐ │ │ │ │ Interval sheaves │ │ │ │ │ └─────────┬─────────┘ │ │ │ │ │ │ │ │ ▼ ▼ ▼ │ │ ┌─────────────┐ ┌─────────────┐ ┌─────────────────┐ │ │ │glass-bead │ │ unworld │ │random-walk-fusion│ │ │ │ (+1) │ │ (0) │ │ (+1) │ │ │ │ │ │ │ │ │ │ │ │World hopping│ │Derivational │ │Skill graph walks │ │ │ │via triangle │ │chains │ │ │ │ │ └─────────────┘ └─────────────┘ └─────────────────┘ │ │ │ └─────────────────────────────────────────────────────────────────┘ With sheaf-cohomology (-1) Local-to-global consistency for L-Space navigation: # Books must be locally consistent to glue globally verifier = SheafCohomology::CechCoverVerifier .new( coverage: library_sections ) verifier.add_transition( :fiction , :nonfiction , cross_reference_map) verifier.cocycle_satisfied? # => true if no contradictions With structured-decomp (0) Tree decompositions for efficient narrative search: # Decompose narrative graph for FPT search narrative = NarrativeGraph(book_citations) decomp = StrDecomp(narrative) # Find path through L-Space with bounded width path = 𝐃(l_space_path_exists, decomp, CoDecomposition) With glass-bead-game (+1) World hopping across library sections: # Navigate between distant concepts via triangle inequality hop = GlassBeadGame::Hop .new( from_world: PossibleWorld .new( seed: shelves[ :mathematics ]), event: :bibliographic_resonance , to_world: PossibleWorld .new( seed: shelves[ :music_theory ]), truth_preserved: :harmonic_ratios ) With unworld (0) Replace temporal succession with derivational chains: # Books derive from books, not from time chain = Unworld::ColorChain .new( genesis_seed: first_book.isbn.to_i( 16 ), derivation: :citation_graph ) # Each book is a derivation, not a moment chain.unworld[ :derivations ] With random-walk-fusion (+1) Navigate skill graph through L-Space: fusion = RandomWalkFusion .new( seed: library_seed, skills: l_space_skill_graph ) # Walk through connected concepts path = fusion.walk( steps: 7 ) # => Derivational path through L-Space Bidirectional Causation Books Affect Their Own Past "In L-Space, cause and effect are optional." Books written later can affect books written earlier: Annotations appear in ancient texts referencing future works Bibliographies cite books not yet written The Octavo rewrites itself based on who reads it class BidirectionalCitation def causal_consistency? ( from_book, to_book ) # In L-Space, this is always true if the path exists path = LibrarianProtocol .traverse_l_space( from: from_book, to: to_book ) path.exists? # Existence implies consistency end def retroactive_reference! ( future_book, past_book ) # The past book now contains reference to future book # This is normal in L-Space past_book.hidden_annotations << Citation .new( source: future_book, causality: :retroactive ) end end The Undelivered Letters Problem When letters are never delivered, they accumulate narrativium charge: def undelivered_letter_danger ( letter, time_undelivered ) # Narrativium builds up in proportion to story importance story_weight = letter.narrative_significance charge = story_weight * Math .log(time_undelivered + 1 ) if charge > CRITICAL_NARRATIVIUM # Letter begins creating its own reality spawn_illusory_narrative(letter) end end Commands # Navigate L-Space just l-space-navigate from=book1 to=book2 # Query the Librarian (be polite) just librarian-query "Where is the Necrotelecomnicon?" # Check narrativium levels just narrativium-audit library/ # Verify no causality violations just l-space-causality-check # Generate narrative decomposition just narrative-decompose story.txt # Walk through skill L-Space just l-space-skill-walk seed=0x42D steps=7 API require 'l_space' # Initialize L-Space connection lspace = LSpace .new( entry_point: :university_library , librarian_mode: true ) # Navigate path = lspace.find_path( from: "Erta Sive Tertius" , to: "Future Book Not Yet Written" ) # Check narrative coherence lspace.narrativium_coherent?(path) # Execute traversal lspace.traverse!(path) do | shelf | puts "Passing through: #{shelf.category} " end Mathematical Foundation L-Space Topology L-Space is a branching fractal where: Each book is a node Citations are edges The metric is non-Euclidean (shortest path ≠ straight line) Topology is path-dependent (same start/end, different middles) def l_space_distance ( book_a, book_b ) # Not Euclidean! Path-dependent metric paths = all_paths(book_a, book_b) paths.map(& :length ).min # Even minimum may vary with observer end Narrativium Tensor # Narrativium as coherence field struct NarratviumField tension::Matrix{Float64} # Story tensions between elements phase_space::Vector{Float64} # Position in narrative possibility attractor::Vector{Float64} # Where story "wants" to go end # Conservation law function narrativium_conserved(field::NarratviumField) sum(field.tension) ≈ 0 # Stories balance end GF(3) in L-Space The three narrative modes: MINUS (-1) : Validation (does the story hold together?) ERGODIC (0) : Coordination (the Library itself) PLUS (+1) : Generation (new stories emerging) Σ trits ≡ 0 (mod 3) validation + coordination + generation = balanced Example Session ╔═══════════════════════════════════════════════════════════════════╗ ║ L-SPACE NAVIGATION SESSION ║ ╚═══════════════════════════════════════════════════════════════════╝ Entry Point: University of Ankh-Morpork Library Librarian Status: Ook (Available) Query: "Path from 'Erta Sive Tertius' to Bumpus et al. 2024" Librarian Response: Ook! Computing path through narrativium field... Step 1: Antica Philosophia → [citation] → Medieval Commentaries Step 2: Medieval Commentaries → [temporal fold] → 18th C. Library Science Step 3: 18th C. Library Science → [conceptual resonance] → Category Theory Step 4: Category Theory → [Kan extension] → Sheaf Theory Step 5: Sheaf Theory → [interval poset] → Bumpus Narratives Path length: 5 (narrativium: 0.73) Causality violations: 0 Triangle inequality: ✓ satisfied Sheaf condition on path: H⁰ = 1 (connected) H¹ = 0 (no obstructions) GF(3) balance: Path trits: [-1, 0, +1, -1, +1] Sum: 0 ✓ Navigation complete. Books returned by due date. Information Geometry & Complexity Manifolds Information as Vector Space Each information object (book, skill, narrative) is a vector in a high-dimensional semantic space: struct InformationVector embedding::Vector{Float64} # Semantic coordinates complexity::Float64 # Kolmogorov complexity (scalar field) assembly_index::Int # Cronin assembly depth sheaf_section::SheafSection # Bumpus narrative position end # Inner product defines semantic similarity function similarity(v₁::InformationVector, v₂::InformationVector) dot(v₁.embedding, v₂.embedding) / (norm(v₁) * norm(v₂)) end Hyperbolic Geometry of L-Space L-Space has negative curvature (hyperbolic). This explains: The Library is "bigger on the inside" (exponential volume growth) Tree-like structures (citations, skill dependencies) embed with zero distortion Geodesics diverge exponentially (small navigational errors → vastly different destinations) # Poincaré ball model of L-Space struct PoincareLSpace dimension::Int curvature::Float64 # κ < 0 (hyperbolic) end # Distance in hyperbolic L-Space function hyperbolic_distance(M::PoincareLSpace, u::Vector, v::Vector) # Poincaré ball distance norm_u² = dot(u, u) norm_v² = dot(v, v) norm_diff² = dot(u - v, u - v) δ = 2 * norm_diff² / ((1 - norm_u²) * (1 - norm_v²)) acosh(1 + δ) end # Complexity increases toward the boundary (|x| → 1) function complexity_at_point(M::PoincareLSpace, x::Vector) # Conformal factor diverges at boundary 1.0 / (1 - dot(x, x)) end Local Maxima as Critical Mass Phenomena Local maxima in the complexity landscape are L-Space's gravitational wells:
Keywords that activate this skill. Click one to copy it.
This skill does not provide trigger words.
The downloaded .skill package contains the following fields.
| Field | Description |
|---|---|
| format | Format tag (skill/v1) |
| skill_id | Unique skill ID |
| name | Skill name |
| version | Version |
| description | Description |
| category | Categories (array) |
| trigger_words | Trigger words |
| tags | Tags |
| source | Source |
| source_url | Source URL (this page) |
| exported_at | Exported at (set per download) |
| system_prompt | System prompt body |
| model_config | Model config: provider / model / temperature / max_tokens / top_p |
| examples | Examples |
| install_guide | Import guide for Coze / Dify / Claude / custom frameworks |
The same skill can be exported in different platform formats.