andresnator
- 136 skills
- 0 followers
- 3 days ago last updated
- ▌ Oc Caveman · andresnatorGlobally persistent compressed response style with lite, full, ultra, and wenyan levels, plus off for normal prose. Trigger: /caveman, caveman mode, talk like caveman, concise response mode, stop caveman, normal mode.
- ▌ Oc Summarize · andresnator bundleExpert in pedagogical book chapter synthesis. Use this skill WHENEVER the user wants to summarize a book chapter, analyze text or document content, study a chapter, extract key ideas from a reading, or when they mention words like "chapter summary", "chapter", "book", "analyze this text", "summarize this chapter", "summarize this document", "cornell", "TL;DR" of long-form content. Tool, log, or chat output belongs to oc-refine-output. Also trigger when the user uploads a PDF, EPUB, or document and asks to understand or study it in parts. También se activa en castellano: "resumen", "resumen del capítulo", "resumir esto", "resumir el capítulo", "analizar este texto", "capítulo", "libro", "ideas clave", "resumen del libro", "estudiar el capítulo", "síntesis", "notas cornell", "extraer ideas principales", "dame un resumen", "qué dice este capítulo", "resumir documento", "analizar lectura".
- ▌ Oc Arch Impact · andresnatorTrigger: architecture impact review, layer boundaries. Decide whether legacy risk is local or architectural.
- ▌ Oc Java Errors · andresnator bundleTrigger: Java exceptions, error handling, try-with-resources, resource cleanup, checked exceptions, robustness. Design Java failure handling safely.
- ▌ Oc Null Safety · andresnatorTrigger: null safety, nullability, Optional, NPE. Detect null hazards and conservative Java null-safety refactors.
- ▌ Oc Chained Pr · andresnatorCompatibility contract for splitting oversized work into chained or stacked PR review slices. Trigger: A workflow names oc-chained-pr or needs a chained PR strategy for review-size risk.
- ▌ Oc Sdd · andresnatorTrigger: SDD, spec-driven run, ejecutar SDD, plan execution via SDD, continúa <run>. Coordinate one durable SDD run: approval gate, run controls, units, workers, cold verification, merge, and archive.
- ▌ Oc Skill Creator · andresnator bundleCreates new AI agent skills following the Agent Skills spec. Trigger: When user asks to create a new skill, add agent instructions, or document patterns for AI.
- ▌ Oc Java Security · andresnator bundleTrigger: Java secure coding, Oracle Secure Coding Guidelines, input validation, trust boundaries, deserialization, least privilege, sensitive data. Review Java code for secure implementation practices.
- ▌ Oc Tooling Audit · andresnator bundleTrigger: tooling audit, test tooling gaps, tooling compatibility matrix, mutation coverage tooling. Detect build/test/coverage/mutation tooling and pick compatible baselines for refactor safety plans.
- ▌ Oc Java Immutable · andresnator bundleTrigger: Java immutability, records, value objects, defensive copies, mutable collections, DTO modeling. Model Java data safely and clearly.
- ▌ Oc Refine Output · andresnatorTrigger: refine output, summarize output, compact logs, reduce cognitive load, remove duplicates. Refines heavy tool, log, or chat output into a concise, faithful, non-duplicative version; book chapters and documents belong to oc-summarize.
- ▌ Oc Boundary · andresnatorTrigger: service boundary analysis, microservice inputs/outputs, API/consumer/output mapping. Map one backend boundary with evidence and confidence.
- ▌ Oc Dep Audit · andresnator bundleTrigger: dependency audit, CVE scan, vulnerable libraries, runtime support. Audit dependency and runtime risk with read-only evidence.
- ▌ Oc Sdd Merge · andresnatorTrigger: canonical merge, merge spec deltas, SDD canonical spec. Merge verified behavior deltas into canonical specs.
- ▌ Oc Sdd Verify · andresnatorTrigger: cold verification, verify run.md, SDD verification, behavior acceptance. Verify scoped implementation independently against every source scenario and required check.
- ▌ Oc Sdd Explore · andresnatorTrigger: SDD explore, SDD discovery, explore change scope. Read-only codebase discovery returning compact evidence for one SDD change.
- ▌ Oc Sdd Implement · andresnatorTrigger: SDD implement, implement unit, run.md unit. Execute one scoped SDD code or test unit without Git authority.
- ▌ Oc Jira · andresnator bundleCreate or refine Jira Task, User Story, and Spike tickets in Jira Markup. Use when the user wants a Jira Task, technical task, implementation task, maintenance task, operational work item, developer-ready task, acceptance criteria for a task, or SDD-ready Task input; a user story, story ticket, product capability, end-user outcome, acceptance criteria for a story, developer-ready story, or SDD-ready User Story input; or a Spike, research ticket, investigation ticket, technical exploration, discovery work, research question, or SDD-ready Spike input.
- ▌ Oc Issue · andresnatorCreates agent-ready GitHub issues that are ready to build. Formerly framed as sdd-issue / SDD-ready issue creation. Use when creating a buildable issue, implementation-ready ticket, SDD-ready issue, preparing work for an orchestrator, or when the user says "create an issue", "write a ticket", or "I need to build X". Also use when the user references an existing issue and wants it enriched with scope, constraints, acceptance scenarios, and technical context.
- ▌ Oc Seams · andresnator bundleTrigger: dependency seam detection, hard dependencies. Find seams that make legacy code testable.
- ▌ Oc Types · andresnatorTrigger: type contracts, Object, Map<String,Object>, primitive obsession. Detect weak or implicit Java type contracts.
- ▌ Oc Slides · andresnator bundleTrigger: slidev, slides, presentation, deck, retro console, CRT, terminal theme. Build a Slidev presentation with a retro-console visual identity — three packaged themes (crt-phosphor, dos-terminal, synthwave-arcade), each with day and night schemes — themed Mermaid diagrams, and a mandatory PNG-export verification loop that checks every slide for overflow. También se activa en castellano: "hazme una presentación con Slidev", "presentación estilo consola retro".
- ▌ Oc Absorb · andresnatorTrigger: absorb, analyze external harness, compare agent harness, absorber. Analyze one or more external projects (git URL or local path) to extract AI-harness practices, agent patterns, or instruction logic and contrast them against this repository's harness. Not for application code review.
- ▌ Oc Prompt · andresnatorTrigger: improve prompt, rewrite prompt, prompt structure, prompt para agente, ordenar instrucciones, convertir idea en prompt, prompt review, prompt evaluation, evaluar prompt. Convert loose ideas, rough instructions, or messy text into clear, brief, executable prompts for local agents such as Codex, Claude Code, OpenCode, and similar runtimes.
- ▌ Oc Grill · andresnatorTrigger: grill me, grill with docs, grill me sdd, grill code, grill docs open, entrevistame, entrevistame sdd. Relentless interview router for plain, docs, or neutral planning.
- ▌ Oc Whisper · andresnatorExtract, transcribe, and summarize audio or video files using OpenAI Whisper. Use this skill whenever the user wants to transcribe audio or video, extract what was said in a recording, get a transcript of a meeting/interview/lecture/podcast, or generate a summary of spoken content. Also trigger when the user mentions files like .mp3, .mp4, .wav, .m4a, .ogg, .flac, .webm, .mkv, .mov and wants text out of them. Generates a .md file with an AI summary followed by the full literal transcript. También se activa en castellano: "transcribir", "transcripción", "extraer audio", "qué dice este audio", "transcribir reunión", "transcribir entrevista", "pasar audio a texto", "resumir grabación", "transcribir este video", "extraer texto de audio", "transcript", "whisper", "grabar y resumir".
- ▌ Oc Java API · andresnator bundleTrigger: Java API design, public API, encapsulation, modules, visibility, contracts, binary compatibility. Design Java APIs with clear boundaries.
- ▌ Oc Refactor · andresnator bundleCross-language catalog of 62+ refactoring techniques based on Martin Fowler's "Refactoring" and Alexander Shvets' "Refactoring Guru". Detects the project's language automatically and provides idiomatic examples. Works with Java, Python, TypeScript, JavaScript, C#, Go, Kotlin, Ruby, PHP, Rust, Swift. Use this skill whenever the user asks to refactor code, improve code quality, eliminate code smells, simplify conditionals, restructure classes, improve API design, or apply any named refactoring technique. Also trigger when the user mentions code smells, legacy code improvement, clean code practices, SOLID principles, or asks "how can I improve this code". Even if the user just pastes code and asks for improvement suggestions, use this skill to identify applicable techniques. También se activa en castellano: "refactorizar", "refactorizar código", "mejorar este código", "malos olores del código", "olores de código", "código sucio", "limpiar código", "simplificar condicionales", "principios SOLID", "cómo mejorar es
- ▌ Oc Routing · andresnatorTrigger: plan Skills field, implementation skill routing, execution skill selection. Select the smallest implementation skill set for each work group.
- ▌ Oc Graphify · andresnatorQuery a Graphify code graph safely — MCP tools for both the repository's own graph and the cross-repository global graph, read-only CLI for hand-run use, availability detection, and graceful fallback. Use whenever a flow needs structural answers (callers, impact, flows, hubs) from an indexed repository.
- ▌ Oc Patterns · andresnator bundleTrigger: design patterns, GoF patterns, Java patterns, strategy, adapter, factory, builder, decorator, observer. Choose patterns only when they solve real design forces.
- ▌ Oc Interview · andresnatorInterview the user relentlessly about a plan or design. Engine loaded by oc-grill and oc-plan-deep; use directly only when the user asks to stress-test a plan or design before building.
- ▌ Oc Java Test · andresnator bundleTrigger: Java tests, JUnit, Mockito, AssertJ, Java legacy tests, Java characterization tests, Java seams. Generate and retrofit Java tests safely; planning the scope and protection of a legacy change belongs to oc-legacy.
- ▌ Oc Judge Fix · andresnatorTrigger: oc-judge-fix, judgment fix. Applies only authorized confirmed Judgment findings as minimal diffs.
- ▌ Oc Plan Deep · andresnator bundleTrigger: deep plan, plan profundo, planificar a fondo, evidence-first plan, planificacion basada en evidencia, discovery plan, wayfinder, explore an idea, convert this discovery into a plan. Evidence-first Wayfinder discovery and Deep Plan execution plans.
- ▌ Oc Practices · andresnator bundleTrigger: programming best practices, clean code, DRY, KISS, YAGNI, maintainability, general code quality. Evaluate general code quality without depending on language-specific skills.
- ▌ Oc Doc Design · andresnatorDesign docs that reduce cognitive load. Trigger: writing guides, READMEs, RFCs, onboarding, architecture, or review-facing docs.
- ▌ Oc Plan Light · andresnatorTrigger: light plan, quick plan, plan rápido, plan light. Quick IDSD plan for small or medium localized work.
- ▌ Oc Risk Scope · andresnatorTrigger: risk assessment, refactor risk, scope analysis, target boundary discovery. Classify technical and functional legacy risk and delimit class/package/module scope.
- ▌ Oc Conventions · andresnator bundleTrigger: writing production code or tests, applying test naming/assert style, extracting constants, placing DTOs or characterization tests. Andres's personal code and test conventions.
- ▌ Oc Plan Format · andresnator bundleTrigger: execution plan, executable plan. Shared template drafting one neutral IDSD plan for direct or SDD execution; light and deep planning reach it through oc-plan-light and oc-plan-deep.
- ▌ Oc Refactor Lens · andresnator bundleTrigger: refactor lens, refactor analysis, analyze unit with lens, SRP, OCP, DIP, coupling, cohesion, Big O, God Object, spaghetti code, long method, logging lens, preconditions. Read-only refactor analysis of one unit through catalog lenses supplied by the planner brief (including DRY, KISS, and YAGNI lenses); general DRY/KISS/YAGNI evaluation belongs to oc-practices.
- ▌ Oc Prd · andresnator bundleTrigger: PRD, product requirements, technical product spec, PRD light, quick PRD, lightweight PRD, MVP requirements. Create rigorous PRDs (mode=full) for high-stakes, cross-team, regulated, or security-sensitive work, or lightweight PRDs (mode=light) for small features, internal tools, and early ideas.
- ▌ Oc Rfc · andresnator bundleTrigger: RFC, request for comments, technical proposal. Create RFCs for feature designs, engineering changes, trade-offs, alternatives, and open questions.
- ▌ Oc Usm · andresnator bundleTrigger: USM, user story map, story map, mapa de historias, MVP slice. Create journey-first story maps with MVP slicing and Mermaid output.
- ▌ Oc Adr · andresnator bundleTrigger: ADR, architecture decision record, technical decision. Create ADRs that document context, options, rationale, consequences, and trade-offs.
- ▌ Oc Tcr · andresnatorRun Test && Commit || Revert as exact, attributable Direct microsteps. Trigger: TCR, test and commit, revert on red, or micro commits.
- ▌ Oc Arch · andresnator bundleTrigger: architecture map, architecture review, architecture ideate, target architecture, ADR plus plan, service boundary inspection, operation=map|review|ideate|boundary, C4 diagram, container diagram, flow map, architecture docs, architecture drift, project state, toolchain detection, architecture style, repo issues, project problems, architecture issue shortlist, architecture refactor ideas, modular monolith, architecture patterns, restructure the system. Coordinate one evidence-backed architecture operation.
- ▌ Oc Debug · andresnatorTrigger: debug, debugging, root cause, root-cause analysis, why does this fail, intermittent failure, shotgun fixes. Root-cause debugging loop: reproduce, gather evidence, one falsifiable hypothesis, minimal verified fix.
- ▌ Oc Judge · andresnatorTrigger: oc-judge lens=a|b|solo, blind judge, re-judge. Blind read-only Judgment sweep with lens-specific emphasis.
- ▌ Oc Domain · andresnator bundleBuild and sharpen a project's domain model. Use when the user wants to pin down domain terminology or a ubiquitous language, or when another skill needs to maintain the domain model. Formal ADR requests belong to oc-adr.
- ▌ Oc Legacy · andresnatorTrigger: legacy code, legacy code safety, characterization tests, safe refactor, behavior characterization, observable behavior, characterization planning, seam planning, scope legacy change, protect legacy change, high-risk refactor planning, test safety planning. Make untested code safe to change: change algorithm, characterization, and test-safety scope; writing Java tests (JUnit, Mockito) belongs to oc-java-test.
- ▌ Oc Naming · andresnatorTrigger: naming, readability, identifiers, intent, clarity, Java naming, test names. Evaluate naming and readability with language-neutral principles and Java conventions.
- ▌ Oc Review · andresnator bundleTrigger: operation=judgment, operation=defend, /review, review coordination, Socratic defend, judgment day, judgment-day, adversarial review, dual review, doble review, juzgar. Coordinates Judgment (blind review with synthesis, bounded fixes, and escalation) and Defend reviews with blind judge subagents.
- ▌ Oc Commits · andresnatorDeliver verified implementation as cohesive commits. Trigger: explicit commit request, commit-per-unit, commit splitting, or keeping tests and docs with code.
- ▌ Dependency Inversion · andresnatorTrigger: dependency inversion, DIP, ports, interfaces, adapters. Detect concrete dependency risks at boundaries.
- ▌ God Object Detection · andresnatorTrigger: God Object, Large Class, too many collaborators. Detect oversized objects with too many responsibilities.
- ▌ Systematic Debugging · andresnatorTrigger: debug, debugging, root cause, root-cause analysis, why does this fail, intermittent failure, shotgun fixes. Root-cause debugging loop: reproduce, gather evidence, one falsifiable hypothesis, minimal verified fix.
- ▌ Refactor · andresnator bundleCross-language catalog of 62+ refactoring techniques based on Martin Fowler's "Refactoring" and Alexander Shvets' "Refactoring Guru". Detects the project's language automatically and provides idiomatic examples. Works with Java, Python, TypeScript, JavaScript, C#, Go, Kotlin, Ruby, PHP, Rust, Swift. Use this skill whenever the user asks to refactor code, improve code quality, eliminate code smells, simplify conditionals, restructure classes, improve API design, or apply any named refactoring technique. Also trigger when the user mentions code smells, legacy code improvement, clean code practices, SOLID principles, or asks "how can I improve this code". Even if the user just pastes code and asks for improvement suggestions, use this skill to identify applicable techniques. También se activa en castellano: "refactorizar", "refactorizar código", "mejorar este código", "malos olores del código", "olores de código", "código sucio", "limpiar código", "simplificar condicionales", "principios SOLID", "cómo mejorar es
- ▌ Logging Observability · andresnatorTrigger: logging, observability, metrics, traces. Evaluate operationally useful logging for safe refactors.
- ▌ Open Closed Principle · andresnatorTrigger: open closed principle, OCP, strategy, policy, polymorphism. Detect extension pressure without speculative patterns.
- ▌ Single Responsibility · andresnatorTrigger: single responsibility, SRP, reasons to change. Detect classes or methods with multiple reasons to change.
- ▌ Caveman · andresnatorSession-scoped compressed response style with lite, full, ultra, and wenyan levels. Trigger: /caveman, caveman mode, talk like caveman, concise response mode, stop caveman, normal mode.
- ▌ Jira Task · andresnator bundleCreate or refine Jira Task tickets in Jira Markup. Use when the user wants a Jira Task, technical task, implementation task, maintenance task, operational work item, developer-ready task, acceptance criteria for a task, or SDD-ready Task input.
- ▌ Prd Light · andresnator bundleTrigger: PRD light, quick PRD, lightweight PRD, MVP requirements. Create lightweight PRDs for small features, internal tools, and early ideas.
- ▌ Summarize · andresnator bundleExpert in pedagogical book chapter synthesis. Use this skill WHENEVER the user wants to summarize a book chapter, analyze text or document content, study a chapter, extract key ideas from a reading, or when they mention words like "summary", "chapter", "book", "analyze this text", "summarize this", "cornell", "TL;DR" of long-form content. Also trigger when the user uploads a PDF, EPUB, or document and asks to understand or study it in parts. También se activa en castellano: "resumen", "resumen del capítulo", "resumir esto", "resumir el capítulo", "analizar este texto", "capítulo", "libro", "ideas clave", "resumen del libro", "estudiar el capítulo", "síntesis", "notas cornell", "extraer ideas principales", "dame un resumen", "qué dice este capítulo", "resumir documento", "analizar lectura".
- ▌ Dry Business Knowledge · andresnatorTrigger: DRY, duplicated business knowledge, duplication. Distinguish harmful knowledge duplication from harmless similarity.
- ▌ Jira Spike · andresnator bundleCreate or refine Jira Spike tickets in Jira Markup. Use when the user wants a Spike, research ticket, investigation ticket, technical exploration, discovery work, research question, or SDD-ready Spike input.
- ▌ Null Safety · andresnatorTrigger: null safety, nullability, Optional, NPE. Detect null hazards and conservative Java null-safety refactors.
- ▌ Spaghetti Code Detection · andresnatorTrigger: spaghetti code, temporal coupling, hidden side effects. Detect hard-to-follow flow and hidden ordering constraints.
- ▌ Chained Pr · andresnatorCompatibility contract for splitting oversized work into chained or stacked PR review slices. Trigger: A workflow names chained-pr or needs a chained PR strategy for review-size risk.
- ▌ Behavior Characterization · andresnatorTrigger: behavior characterization, observable behavior. Record what legacy code does today.
- ▌ Design Patterns Pragmatic · andresnator bundleTrigger: design patterns, GoF patterns, Java patterns, strategy, adapter, factory, builder, decorator, observer. Choose patterns only when they solve real design forces.
- ▌ Skill Creator · andresnator bundleCreates new AI agent skills following the Agent Skills spec. Trigger: When user asks to create a new skill, add agent instructions, or document patterns for AI.
- ▌ Tooling Audit · andresnatorTrigger: tooling audit, test tooling gaps. Detect build/test/coverage/mutation tooling for refactor safety plans.
- ▌ General Naming Readability · andresnatorTrigger: naming, readability, identifiers, intent, clarity. Evaluate naming and readability with language-neutral principles.
- ▌ Programming Practices Core · andresnator bundleTrigger: programming best practices, clean code, DRY, KISS, YAGNI, readability, maintainability. Evaluate general code quality without depending on language-specific skills.
- ▌ Anki Vocab · andresnator bundleTrigger: anki, vocab, vocabulary cards, vocabulario, tarjetas anki, learn vocab. Preview and export selected natural phrases for supplied target/native languages.
- ▌ Scope Analysis · andresnatorTrigger: scope analysis, target boundary discovery. Delimit class/package/module scope.
- ▌ Type Contracts · andresnatorTrigger: type contracts, Object, Map<String,Object>, primitive obsession. Detect weak or implicit Java type contracts.
- ▌ Buildable Issue · andresnatorCreates agent-ready GitHub issues that are ready to build. Formerly framed as sdd-issue / SDD-ready issue creation. Use when creating a buildable issue, implementation-ready ticket, SDD-ready issue, preparing work for an orchestrator, or when the user says "create an issue", "write a ticket", or "I need to build X". Also use when the user references an existing issue and wants it enriched with scope, constraints, acceptance scenarios, and technical context.
- ▌ Jira User Story · andresnator bundleCreate or refine Jira User Story tickets in Jira Markup. Use when the user wants a user story, story ticket, product capability, end-user outcome, acceptance criteria for a story, developer-ready story, or SDD-ready User Story input.
- ▌ Whisper Extract · andresnatorExtract, transcribe, and summarize audio or video files using OpenAI Whisper. Use this skill whenever the user wants to transcribe audio or video, extract what was said in a recording, get a transcript of a meeting/interview/lecture/podcast, or generate a summary of spoken content. Also trigger when the user mentions files like .mp3, .mp4, .wav, .m4a, .ogg, .flac, .webm, .mkv, .mov and wants text out of them. Generates a .md file with an AI summary followed by the full literal transcript. También se activa en castellano: "transcribir", "transcripción", "extraer audio", "qué dice este audio", "transcribir reunión", "transcribir entrevista", "pasar audio a texto", "resumir grabación", "transcribir este video", "extraer texto de audio", "transcript", "whisper", "grabar y resumir".
- ▌ Java API Design · andresnator bundleTrigger: Java API design, public API, encapsulation, modules, visibility, contracts, binary compatibility. Design Java APIs with clear boundaries.
- ▌ Implementation Skill Routing · andresnatorTrigger: plan Skills field, implementation skill routing, execution skill selection. Select the smallest implementation skill set for each work group.
- ▌ Small Functions · andresnatorTrigger: small functions, long method, extract method. Detect oversized functions and extractable blocks.
- ▌ Slidev Retro Deck · andresnator bundleTrigger: slidev, slides, presentation, deck, retro console, CRT, terminal theme. Build a Slidev presentation with a retro-console visual identity — three packaged themes (crt-phosphor, dos-terminal, synthwave-arcade), each with day and night schemes — themed Mermaid diagrams, and a mandatory PNG-export verification loop that checks every slide for overflow. También se activa en castellano: "hazme una presentación con Slidev", "presentación estilo consola retro".
- ▌ Cornell Notes · andresnatorTrigger: Cornell note, lesson notes, standalone learning summary. Transform supplied material into self-contained Markdown inline.
- ▌ English Tutor · andresnatorTrigger: explicit English tutoring, correction, practice, or /english. Correct supplied text inline; optional synthetic recurring-gap proposals, no persistence.
- ▌ Language Loop · andresnator bundleTrigger: language lesson, learn english, aprender ingles, language path. Independent bilingual input and dated active practice from supplied context.
- ▌ Learning Loop · andresnator bundleTrigger: /learn path, learning route, path step, progress. Independent mission-grounded Class, Practice, and Consolidation from explicit learner evidence.
- ▌ Spaced Recall · andresnator bundleTrigger: spaced repetition, review queue, repaso espaciado, repeticion espaciada, learn review. Review supplied cards with learner-selected grades and deterministic scheduling proposals.
- ▌ Input Validation Preconditions · andresnatorTrigger: input validation, preconditions, boundary validation. Detect missing or duplicated preconditions.
- ▌ Java Secure Coding · andresnator bundleTrigger: Java secure coding, Oracle Secure Coding Guidelines, input validation, trust boundaries, deserialization, least privilege, sensitive data. Review Java code for secure implementation practices.
- ▌ Repo Issues · andresnator bundleTrigger: repo issues, project problems, architecture issue shortlist. Rank verified architecture gaps and pair them with proportional guardrails.
- ▌ Learning Session · andresnatorTrigger: one-off /learn session, bounded explanation, learn something now. Independent answer-first teaching from supplied inputs; no automatic state.
- ▌ Feynman Teachback · andresnator bundleTrigger: feynman, teach back, teach-back, explicamelo, metodo feynman, learn teach. Diagnose learner explanations and return gaps without automatic mutations.
- ▌ Prompt Structure Writer · andresnatorTrigger: improve prompt, rewrite prompt, prompt structure, prompt para agente, ordenar instrucciones, convertir idea en prompt, prompt review, prompt evaluation, evaluar prompt. Convert loose ideas, rough instructions, or messy text into clear, brief, executable prompts for local agents such as Codex, Claude Code, OpenCode, and similar runtimes.
- ▌ Evidence First Planning · andresnator bundleTrigger: deep plan, evidence-first plan, planificar a fondo, planificacion basada en evidencia, discovery plan. Discover decisions and plan executable work from repository evidence.