manutej
- 112 skills
- 0 followers
- 63 repo stars
- last month last updated
- ▌
- ▌ Psycopg · manutej bundlePostgreSQL adapter for Python - customer support tech enablement for database operations, query optimization, and data management
- ▌
- ▌ Sheaf Ingest 2 · manutejTurn triples, CSV, wiki markdown, or a JS/TS codebase into a SheafGraph JSON the explorer can load. Use when adding a knowledge graph, scraping a library, validating sheaf JSON, or running from-code / rich-index.
- ▌ Shadcn Motion 2 · manutej bundleUse when generating WOW-level animated components for course pages, dashboards, hero sections, product landing pages, or technical demos. Triggers on 'animated component', 'make this WOW', 'add animation', 'magnetic button', 'number ticker / number flow', 'bento grid', 'hero text reveal', 'hero parallax', 'sticky scroll', 'animated beam', 'dynamic island', '3D tilt card', 'border beam', 'shadcn animations', 'smoothui style', 'aceternity style', 'magic UI', 'motion primitives', 'react bits', 'animate UI', 'cult UI', 'animated dashboard widget', 'scroll storytelling', 'KPI counter', 'animated background', 'interactive experience'. Also use when extending the codebase-to-course design system with motion, building interactive course tutorials beyond single-page HTML, or producing dashboard widgets with data-viz motion. Hybrid shadcn structure + smoothui motion vocabulary.
- ▌ Fp TS 2 · manutej bundleMaster the fp-ts library for typed functional programming in TypeScript, including Option, Either, Task, TaskEither, Reader, State, IO, Array, Record, pipe/flow composition, Do notation, optics (lenses/prisms), and integration with the Effect-TS ecosystem. Use when working with fp-ts data types, composing functional pipelines, handling effects functionally, implementing monadic patterns, or using fp-ts utilities for type-safe functional code.
- ▌ Typescript Fp 2 · manutej bundleMaster functional programming in TypeScript with type-safe patterns, strict typing, advanced type system features, discriminated unions, mapped types, conditional types, and functional patterns. Use when writing TypeScript code with functional paradigms, type-safe error handling with Option/Either types, implementing type-safe composition, leveraging TypeScript's type system for functional patterns, or ensuring compile-time safety in functional code.
- ▌ Functional Programming 2 · manutej bundleMaster functional programming principles and patterns including purity, immutability, composition, higher-order functions, algebraic data types, functors, monads, and effect management. Use when working with functional paradigms, pure functions, immutable data structures, function composition, type-safe error handling, or implementing functional patterns like map/filter/reduce, currying, partial application, recursion, lazy evaluation, or referential transparency.
- ▌ Shadcn Motion · manutej bundleUse when generating WOW-level animated components for course pages, dashboards, hero sections, product landing pages, or technical demos. Triggers on 'animated component', 'make this WOW', 'add animation', 'magnetic button', 'number ticker / number flow', 'bento grid', 'hero text reveal', 'hero parallax', 'sticky scroll', 'animated beam', 'dynamic island', '3D tilt card', 'border beam', 'shadcn animations', 'smoothui style', 'aceternity style', 'magic UI', 'motion primitives', 'react bits', 'animate UI', 'cult UI', 'animated dashboard widget', 'scroll storytelling', 'KPI counter', 'animated background', 'interactive experience'. Also use when extending the codebase-to-course design system with motion, building interactive course tutorials beyond single-page HTML, or producing dashboard widgets with data-viz motion. Hybrid shadcn structure + smoothui motion vocabulary.
- ▌ Sheaf Bind · manutej bundleDomain adapter for sheaf-port. Turn an interconnected source (repo, wiki, knowledge graph, corpus, sensor/dep graph) into a finite cellular sheaf: cells with locators, typed stalks, restriction as field projection, observed incidence only. The only domain-specific surface in the plugin. Use when the user says bind this repo, bind this wiki, instantiate a knowledge sheaf, port the harness onto a domain, or supplies a domain.package.yaml. Trigger on "sheaf-bind", "/sheaf-port", "domain package", "typed stalks", "free-text is not a stalk", "observed incidence".
- ▌ Sheaf Morse · manutejDiscrete Morse reduction of a bound sheaf. Pair cells along invertible, currently-agreeing restrictions; persist the unmatched critical core. Φ is evaluated on the core, not the unreduced cochain. Use when the cover is larger than INNER_BUDGET can glue, or when Tailor must know which cells can actually change Φ. Trigger on "sheaf morse", "morse-reduce", "critical cells", "reduce cochain", "Scythe", "/morse-reduce".
- ▌ Sheaf Ingest · manutejTake a folder, repo, wiki, or knowledge base and build a typed index of its parts, a lattice you can open, and a reusable package so the next folder uses the same structure. Use when the user says "take this codebase and create a sheaf", "build a lattice knowledge base", "ingest this folder", "make a reusable structure from this repo", or "index this wiki." Do not use for a one-file edit or a chat Q&A. Do not treat paragraphs or embeddings as parts. Do not invent links.
- ▌ Sheaf Compose · manutej bundleCompose fleets of fleets operadically — wire independently-coherent teams, sub-agents, or artifact groups together along typed interfaces (an undirected wiring diagram), build the composite coordination sheaf, and find the SEAM: the contradiction that no individual team can see because it does not live inside any team. Use when "compose these teams", "fleet of fleets", "team of teams", "nested agents", "hierarchical orchestration", "wire these subsystems", "the parts all pass but the whole is wrong", "interface contract between subsystems", "where do the chunks join", "seam errors", "split-work-rejoin at scale", "operadic composition", "wiring diagram". Requires one COORD-SHEAF.yaml per box from sheaf-model.
- ▌ Sheaf Enhance · manutejContinue Stalks & Sections enhancement work from HANDOFF.md. Use when implementing views (matrix, slice, multiples, spectral), loader polish, HUD honesty, or example datasets.
- ▌ Meta Mvp · manutej bundleTurn a minimal prompt — even one line — into a complete, WOW-grade deliverable bundle, right-sized to the domain and scope: grounded research, a project vision, a reusable meta-prompt, an execution plan, a grounded data model, and the production-quality artifacts that fit the work (interactive dashboards for a data/pitch/ops product, a report or doc site, an API/spec, a prototype — whatever the domain needs), plus an optional animated explainer with custom SVG. Research-grounded, mobile-ready, and adapts on call to different business/domain needs. Use when the user says meta-mvp, build an MVP or vision or prototype for something, go from this one-liner to something I can show, make me the full pitch, or spin up a new project for a client. Resolves its own scoping by default (never interrogates you — it studies the task's structure and, for high-stakes ambiguity, cross-checks with a small agent panel) and asks only if you opt in. Optionally composes meta-prompting, meta-planning, research-synthesis, user-resea
- ▌
- ▌ Meta Design · manutej bundle> **Meta-Suite doctrine (overrides anything below).** Never ask the user clarifying questions —
- ▌
- ▌
- ▌
- ▌ Meta Planning · manutej bundleGenerate a customized META-PLAN — a typed, example-agnostic planning scaffold — before any complete plan is written, for agentic tasks executed by one or many agents. Bridges /meta-prompting (generic tasks) to agentic planning (ultraplan-class capability). Use whenever the user wants to plan work that agents will execute, orchestrate subagents, decompose a complex build, decide what can run in parallel vs. in sequence, hand context-complete work packets to fresh agents, or asks for "meta-plan", "/meta-planning", "ultraplan", "plan the plan", "deep planning", "map the domain before planning", or "a plan another agent can execute without me". Also use when the input TASK is itself a meta-prompt (operating from meta-meta-prompt space), or when the user wants anti-slop quality gates and kaizen-style iterative execution baked into a plan. Do NOT use for simple single-agent to-do lists a direct plan would serve.
- ▌
- ▌ Meta Prompting · manutej bundle> **Meta-Suite doctrine (overrides anything below).** Never ask the user clarifying questions —
- ▌ Ceti Brand · manutejCanonical CETI brand system shared by noether-course, milton, editorial-dashboard, client-briefing-dashboard, lattice-ops, acos, and client forks (e.g. Hormel). Use whenever building or restyling any CETI dashboard, course shell, briefing, lattice, or when the user mentions CETI brand, CETI tokens, unify dashboards, brand alignment, or Milton × CETI.
- ▌ Sheaf Glue · manutejExact CSP for a capability sheaf. Accept iff every local section is good and every restriction agrees. Relative H¹ is ranking/diagnostic only — never the decision. Analog of meta-review in the sheaf-* family. Use when local skills report success and you need to know whether they form a global section. Trigger on "sheaf glue", "exact CSP", "global section", "Φ(c)", "restriction mismatch".
- ▌ Sheaf Kernel · manutej bundleInstantiate a finite capability sheaf for an agent harness: five vertices (Localization, Contract, Ordering, Preservation, Verification), six restriction maps as field projections, exact CSP Φ as the decision rule, relative cohomology as diagnostic only. Generator of the sheaf-* family — the Batruin analogue of meta-prompting. Use when a harness has multiple components that must agree on shared state, or when sheaf-localize / sheaf-preserve / sheaf-glue / sheaf-repair need a typed incidence graph. Trigger on "capability sheaf", "sheaf kernel", "harness incidence graph", "L C O P V", "restriction atlas".
- ▌ Sheaf Repair · manutej bundleCompositional harness repair via controlled quotients and discrete alignment. Diagnoses restriction obstructions, quotients hidden mediators, re-runs exact CSP. Analog of meta-operad in the sheaf-* family. Zero weight updates: discrete proposals only. Trigger on "sheaf repair", "controlled quotient", "hidden mediator", "stale representative", "align restriction maps".
- ▌ Ceti Research · manutejAuto-research for CETI explainers. Digests a topic into conserved storyboard slots that must glue: metaphor, claim, counterexample, visual motif, duration. Occupies Propose in noether-harness. Never fabricates facts. Use when designing a course film, lookbook page, or ceti-explainer episode. Triggers: auto-research, CETI research, digest topic, storyboard slots, explainer research, /sheaf-run course.
- ▌ Sheaf Harness · manutej bundleOne command that runs the Noether dual-loop with the sheaf-* family as ACV. Tagline: "Glue what the dual-loop conserves." Command: /sheaf-run Automatically Propose (kernel, localize, preserve) → Tailor (glue ⇄ repair) → Commit (Φ) → Meta (discrete only). Use whenever the user wants the final harness, a workflow YAML, or a simple slash command instead of calling sheaf skills by hand. Triggers: /sheaf-run, /sheaf, run the harness, glue what the dual-loop conserves, sheaf harness, noether sheaf workflow.
- ▌ Ceti Explainer · manutej bundleBuild a gold-standard animated technical explainer: one moving SVG diagram that builds a complex concept step by step with real numbers, in a video-like player (scrub, captions, chapters, keyboard) on a deterministic clock so every frame is reproducible. Output is a single self-contained HTML file that works offline. Use when the user wants to explain how something works as a smooth, followable animation — ML/AI internals (attention, embeddings, RAG, MCP), systems and protocols (TCP, OAuth), algorithms, data structures, or any interconnected technical flow. Triggers: animated explainer, explainer video, explain X visually, show how X works step by step, animated diagram of X. Two formats: the ~40s 8-beat episode (default) and the ~2-min 7-movement feature cut (LONGFORM.md) — for video requests, lay audiences, or broad topics. Not for marketing pages, scroll-driven essays, slide decks, or static posters.
- ▌ Sheaf Localize · manutejProduce Localization and Contract stalks for a capability sheaf: path, symbol, namespace, API revision, source commit as typed signatures. Binds to sheaf-kernel. Analog of meta-planning in the sheaf-* family. Use when a harness repair needs typed “where” and “what interface” instead of free-text retrieval. Trigger on "sheaf localize", "localization stalk", "contract stalk", "typed file+API signature".
- ▌ Sheaf Preserve · manutejProduce Ordering and Preservation stalks: edit order, test id, commit, namespace, path — what must not break, in what sequence. Binds to sheaf-kernel. Analog of meta-design in the sheaf-* family. Use when edits must keep public API, tests, and module boundaries. Trigger on "sheaf preserve", "ordering stalk", "preservation stalk", "edit atoms sheaf".
- ▌ Noether Harness · manutej bundleGenerate conservation-constrained agentic coding harnesses from the Noether Dual-Loop kernel — a μ meta-meta-prompt that emits a λ meta-prompt, which emits a concrete harness (typed conservation lens, four role prompts, per-framework fragments) for LangGraph, Aider, Cursor, OpenHands, AutoGen, ReAct+tools, or custom loops. Use whenever the user wants a self-refining coding agent where tests, type-checks, API stability, and architecture act as conserved invariants; a propose→tailor→commit→reflect dual loop with a bounded inner refinement loop and an outer invariant-learning loop; prediction-time "tailoring" (Alet et al.) transposed to LLM agents; or adaptation with ZERO weight updates / no fine-tuning. Also use to instantiate, extend, or explain such a harness. Trigger on "Noether harness", "conservation lens", "dual-loop agent", "tailoring loop", "tests-as-invariants harness", or a reference to the Noether meta-meta-prompt — even phrased only as "stop my agent breaking tests when it refactors."
- ▌ Code Smells · manutejDetection lens for Fowler's five smell families — names the smell, states the change-cost, and hands the fix to `refactoring-patterns`. Use when reviewing a diff, auditing a module, or deciding whether to gate a merge. Triggers: "does this code have smells", "review for smells", "is this clean", "what's wrong with this design", "this class is getting big", "why is every change touching 12 files", "switch on type", "this method is too long", "is this AI slop".
- ▌ Dry And Reuse · manutejStops duplication and reinvention — the failure GitClear measures as "rework" (copy-paste up, refactoring down, the signature of AI code with a short shelf life). Use before writing any helper, util, type, constant, or component, and when reviewing a diff for repeated logic. Enforces search-before-build, consolidate- don't-clone, and DRY-as-knowledge (one rule, one authoritative place). Triggers: "is there already a function for this", "this looks duplicated", "DRY this up", "extract the common logic", "we keep rewriting this", before adding any utility.
- ▌ Production Grade · manutejRouter and orchestrator for the craft code-quality guardrails. Use FIRST on any broad "is this production-ready?", "review this code/diff/PR", "build this properly", "why does our AI-generated code keep needing rework", or "make this maintainable" request. Two modes: GENERATIVE GUARD (load the right senior heuristics before writing code) and CRITIQUE (audit a diff/file and return one worst-first ranked list of findings). Decomposes the request, fires the relevant craft member skills plus the existing complexity/refactoring/dependency/test skills in the right order, resolves their conflicts, and returns a single merged verdict. Defer to a single member skill only when the task is already narrow.
- ▌ Trustworthy Tests · manutejMakes tests prove something and makes "done" mean done. Use when writing tests, judging existing tests, or before claiming a task is complete/production-ready. Catches tautological tests (asserting on their own mocks), tests that mirror the implementation, fake green CI, and the agent's habit of declaring victory without ever running the code. Enforces test-observable-behavior, the verification-before- done evidence gate, and characterization-tests-before-refactoring-untested-code. Triggers: "write tests for this", "are these tests any good", "is this done", "is it production-ready", "add coverage", before any "done"/handoff claim.
- ▌ Effects And Purity · manutejSeparates pure decision logic from I/O so code is testable without heavy mocks — without demanding religious purity. Use when a function both computes and writes to a DB, network, or log; when datetime.now()/random()/fetch() are called inline inside business logic; when a unit test needs five mocks to run; when reviewing code that tangles "what to do" with "do it." Triggers: "this is hard to test," "I need to mock the clock," "how do I test this without a DB," "pure function," "side effects," "functional core," "inject the dependency." NOT a trigger for: making everything pure, banning mutation, or wrapping simple code in monads.
- ▌ Right Sized Design · manutej bundleFights over-engineering and premature abstraction — the #1 documented AI coding failure. Use when about to add a class, interface, factory, base-class, config layer, plugin system, generic "framework," or any abstraction; when a diff feels bigger than the request; or when reviewing code that has more structure than the problem warrants. Enforces smallest-diff-that-works, YAGNI, "no abstraction without 2-3 real call sites," and Ousterhout's deep-vs-shallow-module test. Triggers: "is this over-engineered", "simplify this", "do we need this abstraction", "this feels heavy", "make it production-ready" (often means add-robustness-not-structure).
- ▌ Naming And Comments · manutejNames and comments are the primary channel through which code communicates intent to the next engineer. Use when choosing an identifier ("what should I call this", "rename this", "is this name clear"), deciding whether to add a comment ("should I comment this", "does this need explaining"), or auditing existing prose ("clean up the comments", "too many comments", "comment slop"). Also fires when a name is hard to pick — that difficulty is a design smell, not a vocabulary problem (Ousterhout: Hard to Pick Name).
- ▌ Supply Chain Hygiene · manutejThe verification gate AI agents skip. Use whenever you add a dependency, ask "is this package safe", import a library, call an external API, pin versions, or check a package exists. Catches hallucinated packages (19.7% of AI suggestions, USENIX 2025), hallucinated APIs, typosquats, and inlined secrets before they reach the codebase. Triggers: "add a dependency", "is this package safe", "import this library", "call this API", "pin versions", "check this package exists", any new third-party import.
- ▌ Data And State Evolution · manutej bundleMakes schema, migration, and data-shape changes safe to ship. Use when writing a database migration, altering a table, adding/renaming/dropping a column, changing the shape of an API payload / event / serialized object, backfilling data, or reviewing any diff that touches persistent state. Catches one-shot breaking changes, NOT NULL without backfill, missing/untested rollbacks, nullability drift, and backfills that lock production tables. Enforces expand→migrate→ contract, one-deploy-window backward compatibility, and dry-run-on-a-copy evidence. Triggers: "write a migration", "alter table", "add a column", "rename this field", "change the schema", "is this migration safe", "change this response/event shape", "backfill".
- ▌ Robustness At Boundaries · manutej bundleThe robustness the AI skips. Use whenever code reads untrusted input (API request, user input, file, env, parsed JSON, DB row, external response), handles errors, or touches secrets/config. Enforces validate-at-the-edge, never-swallow-errors, define-errors-out-of-existence, no-hardcoded-secrets, and meaningful error contracts. Counters the #1 AI omission: agents over-build structure but skip input validation and error handling exactly where production breaks. Triggers: "add error handling", "validate this input", "handle the edge cases", "is this safe", "production hardening", any try/catch, any parse/deserialize, any API handler.
- ▌ Hunt · manutej bundleHunts a prospect's provable weaknesses and buying tells from public sources and packages them as a typed, cited tells-dossier — the left half of the sign-here bijection (tells ↔ scope ↔ price) that makes the pitch feel painfully obvious. Use when the user says "hunt [company]", "find their tells", "weakness map", "recon this prospect", "why would they buy", "what are their gaps", "build the dossier", or before any bid, deck, demo, or outreach so positioning rests on demonstrated gaps instead of generic value props. Runs one prospect or fans out across many. Produces a cited dossier with severity/provability/monetizability scores, a live-demo wedge list, a quarantined-hypotheses section, and a NEEDS/UNKNOWNS handoff — never client-facing.
- ▌
- ▌ Demo Forge · manutej bundleBuilds the touchable sample that enacts the buyer's own workflow step — an interactive single-file prototype, a ceti-method animated explainer, or a live-channel (phone/chat) demo — so the buyer experiences the offer instead of reading about it, with buildable per-form artifact specs and a rehearsed 3-minute run-of-show. Use when the user says "demo for [prospect]", "build the sample", "prototype they can touch", "demo plan", "show don't tell", or as the 04-demo stage of a factory run. A demo the buyer can call or click beats any document; a generic vertical demo demonstrates nothing.
- ▌ Design Dna · manutej bundleExtracts a prospect's real brand palette from their public materials, harmonizes it into a WCAG 2.2 AA-audited token set, and fuses it with CETI house rigor so every deliverable feels already theirs. Emits design/tokens.css (the versioned sf-tokens schema every artifact skill consumes) and design/palette.md (provenance + contrast matrix). Use when the user says "design DNA for [company]", "extract their colors", "brand the deliverables for this prospect", "tokens for this client", or automatically as the design stage of any factory run before artifacts are built. Refuses to ship invented colors or failing contrast pairs.
- ▌ Mint Atoms · manutej bundleThe compounding engine of signal-forge — captures reusable knowledge atoms from every run and every human correction, traces each correction to the exact factory line that allowed it, writes append-only ledgers with a same-transaction discoverability index, and refreshes decay-prone atoms on a cadence so run
- ▌ Pitch Deck · manutej bundleBuilds the proposal slide deck for a prospect — an animated single-file HTML deck by default, .pptx when the buyer's meeting culture demands Office, or both — walking from evidenced weaknesses (tells) to the sign-here moment. Use when the user says "pitch deck", "proposal deck", "slides for [prospect]", "deck for the meeting", or as the 05-deck stage of a factory run. Every number comes from SYNTHESIS; every scope item traces to a tell; the tells ↔ scope ↔ price bijection is the spine. The HTML artifact is a precise, buildable spec (16:9 stage, anime.js v4, `--sf-*` tokens, ≤2.5MB, zero console errors, WCAG AA, prefers-reduced-motion) — not a vibe.
- ▌ Run Factory · manutej bundleExecute a deliverable factory end-to-end on one prospect signal, or a batch of them — intake, digest, parallel cited research lanes, weakness hunt, synthesis, adversarial citation audit, then fan out the winning artifacts (dashboard, bid, deck, demo, copy). Use when the user pastes a prospect signal (text, email, call note), says "run the factory", "new prospect", "bid package for X", "full pipeline on this lead", "portfolio run" or "run these N leads", "resume the run", or names a registered factory to execute. This is the flagship orchestration harness of signal-forge.
- ▌ Account Copy · manutej bundleEmits reader-specific account copy — outreach sequences, follow-ups, one-pagers, case-study atoms, QBR narratives, renewal and expansion notes — by selecting a typed sequence blueprint and filling its slots only with facts the account already holds, so every piece reads like it was written for one named reader and never like mail-merge. Use when the user says "outreach for [prospect]", "follow-up email", "nurture sequence", "one-pager", "case study", "QBR narrative", "renewal note", "win-back", "expansion copy", or when a factory contract includes 07-copy. Runs one reader or batches a whole list with per-reader verification. Every claim cites SYNTHESIS or is cut; refuses unpersonalizable sends and "just checking in" revivals.
- ▌
- ▌
- ▌ Purify · manutej bundleMaster the Purify library for practical functional programming in TypeScript with algebraic data types (Maybe, Either, EitherAsync, MaybeAsync), composable error handling, data transformations, Codec for runtime type safety, List operations, Tuple utilities, and functional patterns. Use when working with Purify's Maybe/Either types, async error handling with EitherAsync/MaybeAsync, runtime validation with Codecs, or building practical functional TypeScript applications with cleaner syntax than fp-ts.
- ▌
- ▌
- ▌
- ▌
- ▌
- ▌
- ▌
- ▌
- ▌
- ▌ Meta Self · manutej bundleMaster reference for categorical meta-prompting unified syntax. Contains all modifiers, operators, composition patterns, and execution protocols. Use this skill for self-reference when executing any prompt workflow, ensuring consistent syntax across all commands and skills.
- ▌ Prompt Dsl · manutej bundleDomain-specific language for categorical prompt composition with functor combinators and natural transformation operators. Use when building composable prompt templates, implementing typed prompt algebras, creating reusable prompt patterns with categorical structure, or designing prompt DSLs that preserve composition properties.
- ▌ Discopy Nlp · manutej bundleDisCoPy categorical quantum NLP for string diagrams, monoidal categories, and compositional semantics in Python. Use when implementing compositional distributional semantics, building string diagram computations, working with monoidal category theory in code, creating quantum-inspired NLP models, or applying categorical semantics to natural language processing.
- ▌ Llm4s Scala · manutej bundleLLM4S Scala functional LLM interfaces with Effect system integration. Use when building LLM applications in Scala with ZIO or Cats Effect, implementing type-safe AI pipelines with functional error handling, creating composable prompt systems in Scala, or leveraging Scala's type system for robust AI applications.
- ▌ Effect TS AI · manutej bundle@effect/ai integration patterns for categorical AI composition, typed error handling, and production prompt pipelines. Use when building AI applications with Effect-TS, composing LLM calls with typed errors, creating tool-augmented AI systems, implementing structured output generation, or integrating multiple AI providers (OpenAI, Anthropic) with categorical composition patterns.
- ▌ Atomic Blocks · manutej bundleComposable atomic blocks for categorical meta-prompting. Defines 16 reusable blocks across 4 layers (Assessment, Transformation, Refinement, Composition) that can be composed to build custom workflows. Enables progressive disclosure: simple commands work unchanged, power users can override individual blocks or compose freely.
- ▌ Hasktorch Typed · manutej bundleHasktorch type-safe tensor operations with categorical structure preservation. Use when implementing type-safe deep learning in Haskell, leveraging dependent types for tensor shape verification, applying categorical abstractions to neural network design, or building formally verified ML pipelines with strong type guarantees.
- ▌ MCP Categorical · manutej bundleMCP (Model Context Protocol) server patterns with categorical tool composition and typed context protocols. Use when building MCP servers with categorical composition patterns, designing tool interfaces using functors and natural transformations, implementing typed context management, or creating composable AI tool ecosystems with categorical guarantees.
- ▌ Dspy Categorical · manutej bundleDSPy compositional prompt optimization with categorical signatures, module chaining, and automated prompt tuning. Use when building declarative LLM programs with typed signatures, composing multi-step reasoning modules (ChainOfThought, ReAct, ProgramOfThought), optimizing prompts with MIPROv2/BootstrapFewShot, or creating modular AI pipelines that separate program logic from prompt engineering.
- ▌ Lmql Constraints · manutej bundleLMQL constraint-guided generation DSL for type-safe prompting with grammar constraints and logical conditions. Use when building structured LLM outputs with guaranteed format compliance, implementing constrained decoding with logical operators, creating type-safe prompt templates, or combining neural generation with symbolic constraints for reliable AI outputs.
- ▌ Prompt Benchmark · manutej bundleSystematic prompt evaluation framework with MATH, GSM8K, and Game of 24 benchmarks. Use when evaluating prompt effectiveness on standard benchmarks, comparing meta-prompting strategies quantitatively, measuring prompt quality improvements, or validating categorical prompt optimizations against ground truth datasets.
- ▌ Guidance Grammars · manutej bundleGuidance library grammar-constrained generation with categorical structure. Use when implementing structured LLM outputs with grammar constraints, building template-based generation with guaranteed formats, creating type-safe prompt templates with interleaved computation, or applying categorical grammar theory to constrained generation.
- ▌ Polynomial Functors · manutej bundleSpivak-Niu polynomial functor implementations for learner composition and dynamical systems. Use when modeling learning systems as categorical structures, composing machine learning components with polynomial functors, implementing the categorical framework from Polynomial Functors - A Mathematical Theory of Interaction, or building compositional dynamical systems with lenses and charts.
- ▌ Arxiv Categorical AI · manutej bundleSystematic analysis patterns for categorical AI papers (Gavranović, de Wynter, Bradley, Zhang). Use when analyzing academic papers on categorical deep learning, extracting categorical structures from AI research, mapping meta-prompting concepts to category theory, or building comprehensive literature reviews on categorical approaches to AI.
- ▌ Voltagent Multiagent · manutej bundleVoltAgent multi-agent system design with natural transformation coordination between agents. Use when building TypeScript multi-agent AI systems, implementing agent coordination with categorical patterns, designing supervisor-worker agent hierarchies, or creating composable agent architectures with typed message passing.
- ▌ Cc2 Research Framework · manutej bundleCC2.0 seven-function research workflow (observe, reason, create, orchestrate, learn, verify, deploy) for categorical AI research. Use when conducting systematic research on categorical AI topics, coordinating multi-stream research workflows, applying categorical foundations to research methodology, or implementing the L5 meta-prompting research framework.
- ▌ Dynamic Prompt Registry · manutej bundleDynamic prompt registry with unified categorical syntax. Supports @skills:discover(), @skills:compose(A⊗B), Reader monad for runtime lookup, and quality-tracked prompt libraries. Use for meta-prompts referencing sub-prompts, building prompt libraries, implementing deferred resolution, or composing templates dynamically.
- ▌ Langgraph Orchestration · manutej bundleLangGraph stateful multi-agent graphs with categorical coordination patterns and cyclic workflows. Use when building stateful AI agent systems, implementing multi-agent orchestration with conditional routing, creating cyclic workflows with persistence, or designing graph-based AI pipelines with checkpointing and human-in-the-loop patterns.
- ▌ Recursive Meta Prompting · manutej bundleRecursive Meta-Prompting (RMP) implementation with unified categorical syntax. Supports @mode:iterative, @quality: thresholds, >=> Kleisli composition, and comonadic context extraction. Use when implementing iterative prompt improvement, quality-gated generation loops, or applying categorical fixed-point semantics with convergence guarantees.
- ▌ Categorical Meta Prompting · manutej bundleFull categorical framework for meta-prompting with verified laws. Use when implementing functor-based task routing, monadic iterative refinement, comonadic context extraction, or quality-tracked composition. Integrates F (Functor), M (Monad), W (Comonad), and [0,1]-enriched quality categories.
- ▌ Quality Enriched Prompting · manutej bundle[0,1]-enriched category implementation for gradient-based prompt quality optimization. Use when implementing quality-aware prompt systems, building enriched categorical structures for prompt evaluation, creating continuous optimization over prompt spaces, or applying Bradley's enriched category theory to language model quality scoring.
- ▌ Categorical Property Testing · manutej bundleProperty-based testing for functor laws, monad laws, and naturality conditions using fp-ts and fast-check. Use when validating categorical implementations in TypeScript, testing algebraic laws in functional code, verifying functor/monad/applicative instances, or building test suites for categorical abstractions.
- ▌ Categorical Structure Builder · manutej bundleUniversal template for implementing categorical structures in meta-prompting frameworks. Applies to functors, monads, comonads, natural transformations, adjunctions, hom-equivalences, limits, colimits, and enriched categories. Use when extending frameworks with new categorical constructs.
- ▌ Discopy Categorical Computing · manutej bundleCategory theory for compositional computing with string diagrams, quantum circuits, and QNLP. Covers monoidal categories, functors, tensor evaluation, and practical applications in quantum natural language processing and diagrammatic reasoning.
- ▌ Fp TS · manutej bundleMaster the fp-ts library for typed functional programming in TypeScript, including Option, Either, Task, TaskEither, Reader, State, IO, Array, Record, pipe/flow composition, Do notation, optics (lenses/prisms), and integration with the Effect-TS ecosystem. Use when working with fp-ts data types, composing functional pipelines, handling effects functionally, implementing monadic patterns, or using fp-ts utilities for type-safe functional code.
- ▌ Typescript Fp · manutej bundleMaster functional programming in TypeScript with type-safe patterns, strict typing, advanced type system features, discriminated unions, mapped types, conditional types, and functional patterns. Use when writing TypeScript code with functional paradigms, type-safe error handling with Option/Either types, implementing type-safe composition, leveraging TypeScript's type system for functional patterns, or ensuring compile-time safety in functional code.
- ▌ Functional Programming · manutej bundleMaster functional programming principles and patterns including purity, immutability, composition, higher-order functions, algebraic data types, functors, monads, and effect management. Use when working with functional paradigms, pure functions, immutable data structures, function composition, type-safe error handling, or implementing functional patterns like map/filter/reduce, currying, partial application, recursion, lazy evaluation, or referential transparency.
- ▌ MCP Integration Expert · manutej bundleExpert guidance for researching, documenting, and integrating Model Context Protocol (MCP) servers and tools. Covers MCP architecture, server/client implementation patterns, tool discovery, integration workflows, security best practices, and multi-language SDK usage (Python, TypeScript, C#, Java, Rust). Enables seamless integration of MCP tools into Claude Code and AI applications.
- ▌ Pytest Patterns · manutej bundlePython testing with pytest covering fixtures, parametrization, mocking, and test organization for reliable test suites
- ▌ Redis State Management · manutej bundleComprehensive guide for Redis state management including caching strategies, session management, pub/sub patterns, distributed locks, and data structures
- ▌ CI CD Pipeline Patterns · manutej bundleComprehensive CI/CD pipeline patterns skill covering GitHub Actions, workflows, automation, testing, deployment strategies, and release management for modern software delivery
- ▌ Observability Monitoring · manutej bundleComprehensive observability and monitoring skill covering Prometheus, Grafana, metrics collection, alerting, exporters, PromQL, and production monitoring patterns for distributed systems and cloud-native applications
- ▌ Terraform Infrastructure · manutej bundleComprehensive Terraform infrastructure-as-code skill covering providers, resources, modules, state management, and enterprise patterns for multi-cloud infrastructure
- ▌ React Development · manutej bundleComprehensive React development with hooks, components, state management, context, effects, and performance optimization based on official React documentation
- ▌ Responsive Design · manutej bundleMobile-first responsive design covering fluid layouts, media queries, flexbox, grid, viewport units, and responsive images
- ▌ Langchain Orchestration · manutej bundleComprehensive guide for building production-grade LLM applications using LangChain's chains, agents, memory systems, RAG patterns, and advanced orchestration
- ▌ Nextjs Development · manutej bundleComprehensive Next.js development skill covering App Router, Server Components, data fetching, routing patterns, API routes, middleware, and full-stack Next.js applications