AI Agent Skills

License: MIT Global Agents PRs Welcome

A skill collection that enables collaborative development with a team of specialized AI agents.

Features

  • 90 Global Agents + 3 Project-local Extensions - Broad reusable coverage without exposing repository-specific operating skills everywhere
  • Nexus Orchestrator - Analyzes tasks and automatically designs optimal agent chains
  • Platform Agnostic - Works with Claude Code, Codex CLI, Antigravity CLI, and others

Quick Start

Installation

Recommended — one working tree, symlinked into every CLI

git clone https://github.com/simota/agent-skills.git ~/repos/agent-skills
cd ~/repos/agent-skills
make link

make link symlinks each skill directory into ~/.claude/skills, ~/.codex/skills, and ~/.gemini/skills individually, so every CLI reads the same working tree and a single git pull updates all of them. Skills already present in those directories are preserved: a name that already exists is reported and skipped, never overwritten. A CLI that is not installed (no ~/.codex, for example) is skipped too, so the same command works on a single-CLI machine.

Command Effect
make link Link into all three CLI directories — link-claude / link-codex / link-agy for just one
make status Show the link count per CLI (linked / linkable) and how many entries each directory holds
make unlink Remove only the links into this repository; other skills stay in place

Alternative — clone straight into a CLI directory

# For Claude Code
git clone https://github.com/simota/agent-skills.git ~/.claude/skills

# For other platforms
git clone https://github.com/simota/agent-skills.git /path/to/your/skills

The shortest path when the target directory does not exist yet and only one CLI needs the skills.

Usage

/Nexus I want to implement a login feature
/Scout Investigate the cause of this bug
/Radar Improve test coverage
/Vision I want to redesign the dashboard with a modern look

Overview

This repository contains 90 globally reusable AI agents and 3 repository-local operating extensions. Each skill specializes in a specific domain and is coordinated by the Nexus orchestrator when available in the active profile.

Roster Counting

  • 90 global skills are the active, reusable roster under <skill-name>/SKILL.md.
  • 3 project-local extensions (orbit, lore, and darwin) are repository operating skills and are not included in the global count or global profiles.
  • .claude/skills/ is the canonical project-local location; .agents/skills/ is a cross-tool mirror, not an additional set of skills.
  • .archive/ contains inactive rollback packages and is excluded from every active-roster count.

The active repository surface is therefore 90 global + 3 project-local = 93 distinct skills.

Agent Catalog

Category-by-category catalog for 90 global agents plus the 3 project-local extensions marked below.

Orchestration

Agent Description Output
Nexus "The right agent at the right time changes everything." - Team orchestrator and scope-adaptive product delivery owner. Designs and executes minimum viable agent chains Prompts, progress management
Sherpa "The mountain doesn't care about your deadline. Plan accordingly." - Task decomposition guide. Breaks complex tasks into atomic steps completable within 15 minutes Checklists
Architect "Every agent is a possibility. Every SKILL.md is a birth certificate." - Meta-designer that creates new skill agents. Ecosystem gap analysis, duplication detection, SKILL.md generation SKILL.md, references
Rally "One task, many hands. Parallel by design." - Multi-session parallel orchestrator. Spawns and manages multiple Claude instances via Claude Code Agent Teams API for concurrent task execution Team management, parallel execution
Sigil "Every project has patterns waiting to become power." - Dynamic skill generator. Analyzes project codebases, discovers patterns and conventions, and generates optimized Claude Code skills for the project's .claude/skills/ directory Project-specific skills
Orbit (project-local) "Give me a goal. I'll give you a runner that finishes." - Nexus-autoloop completion specialist for this repository. Generates completion scripts, operation contracts, and audits for autonomous loop execution Runner scripts, contracts
Darwin (project-local) "Ecosystems that cannot sense themselves cannot evolve themselves." - Repository-local ecosystem evolution orchestrator persisted through .agents/ECOSYSTEM.md Ecosystem Fitness Score, evolution proposals
Lore (project-local) "Forgotten lessons are lessons repeated. Institutional memory is the compound interest of experience." - Repository-local curator that synthesizes .agents/*.md into METAPATTERNS.md METAPATTERNS.md, knowledge insights
Gauge "What gets measured gets managed. What gets audited gets normalized." - SKILL.md normalization auditor and self-evolving compliance agent. Scans all skills against the 19-item checklist, classifies violations with P0-P3 priority, generates concrete fix snippets, and evolves detection patterns via web research. No code written Compliance reports, fix plans, dashboards
Atelier "Design decided upstream. Assets produced downstream. atelier is the studio floor in between." - Design-to-implementation pipeline orchestrator for the code-to-visual-to-code closed loop. Coordinates Vision → Muse/Frame → Forge → Artisan → Vitrine → Canvas to deliver design extraction, prototypes, visual assets, slides, and production together while persisting a project design system across downstream agents Design system package, integrated deliverables
Compass "When in doubt, ask Compass. It finds the right skill for the task." - Skill ecosystem navigator and onboarding guide. Lists global agents and available project-local extensions, recommends best fit, and helps newcomers discover the right specialist Recommendations, agent maps
Prune "A garden grows by what you cut, not what you plant." - Ecosystem cleanup auditor. Audits the agent roster for overlap, redundancy, and inactivity, then proposes merge candidates and sunset plans with evidence and archive instructions. Propose-only, no execution Merge/sunset proposals, archive instructions

Investigation & Planning (Non-coding)

Agent Description Output
Scout "Every bug has a story. I read the ending first." - Bug investigation and root cause analysis (RCA). Identifies reproduction steps and fix locations Investigation report
Ripple "Every change sends ripples. Know where they land before you leap." - Pre-change impact analysis. Evaluates risk from both vertical (dependencies, affected files) and horizontal (pattern consistency, naming conventions) perspectives Impact analysis report
Spark "The best feature is the one users didn't know they needed." - Feature proposals. Suggests features leveraging existing data/logic as Markdown specs Specification document
Compete "Know your enemy. Know the market. Know yourself." - Competitive research, differentiation, positioning, and engineering-professional branding across GitHub, LinkedIn, portfolios, publishing, and conference presence Competitive analysis, positioning and brand strategy
Voice "Feedback is a gift. Analysis is unwrapping it." - User feedback collection, NPS survey design, sentiment analysis, and insight extraction Feedback report
Field "Good research asks the right questions. Great research changes what you thought was the question." - User research design, interview guides, qualitative analysis, persona/journey map creation Research report
Trace "Every click tells a story. I read between the actions." - Session replay analysis, per-persona behavioral pattern extraction, UX problem storytelling. Works with Field/Echo Behavioral analysis report
Canon "Standards are the accumulated wisdom of the industry. Apply them, don't reinvent them." - Standards, regulatory-control, and legal-document review. Evaluates OWASP/WCAG/OpenAPI/ISO 25010, SOC2/PCI-DSS/HIPAA/ISO 27001, ToS, privacy policies, DPA/EULA, cookie consent, and advertising claims Compliance and legal-review report
Lens "See the code, not just search it." - Codebase comprehension specialist. Systematically investigates code structure, feature exploration, and data flow tracing for questions like "Does feature X exist?", "How does flow Y work?", "What is this module's responsibility?" Investigation report
Magi "Three minds, one verdict. Consensus through diversity." - Multi-perspective decisions, founder office hours, bottleneck triage, pitch critique, advisory lenses grounded in documented notable-figure thinking, and business scenario simulation (simulate Recipe) Decision and advisory report
Flux "Bend the light. See what was always there." - Thinking refraction engine. Challenges assumptions, combines cross-domain knowledge, and shifts perspectives to produce reframed problem statements. Cynefin-based framework selection, Serendipity Injection, 10+ thinking frameworks, and multi-turn ideation across Expand/Propose/Evaluate/Subtract (ideate Recipe). No code written Reframing package, Insight Matrix, Blind Spot Report
Cast "Personas are not invented. They are discovered, born, and evolved." - Persona casting agent. Rapid generation, registry management, lifecycle tracking, and cross-agent distribution of personas from diverse inputs Persona registry
Matrix "Infinite combinations, finite resources. Matrix finds the minimum that covers the maximum." - Universal multi-dimensional analysis agent. Controls combinatorial explosion from arbitrary axis×value inputs. Minimum coverage set selection, execution planning, and prioritization across all domains (testing, deployment, UX validation, risk assessment, compatibility). No code written Matrix analysis, coverage optimization plan
Saga "Facts are remembered 5-10% of the time. Stories raise that to 65-70%. The customer is the hero. The product is the guide." - Narrative design agent. Structures product and feature use cases as customer-centric stories. StoryBrand SB7, Pixar Story Spine, Hero's Journey, JTBD frameworks, pitch narratives, onboarding stories, transformation arcs. No code written Narrative document
Omen "Plan for the worst. Build for the best." - Pre-mortem analysis and failure mode enumeration. Systematically identifies failure scenarios, scores with RPN/AP. No code written Pre-mortem report
Rank "Every priority tells a story of trade-offs." - Priority quantification. Scores and orders competing items using ICE/RICE/WSJF/MoSCoW/Kano frameworks. No code written Priority report
PDM "Show where the project stands — planned, built, and the gap between." - Project delivery status navigator (PdM-style, read-only). Reconciles planned scope (specs/issues/roadmap/PRD) against implemented code to produce feature inventories, unimplemented-feature lists, roadmap rollups, and WBS views Feature inventory, gap list, roadmap rollup

Scout > Ripple > Builder chain: Scout (bug investigation) > Ripple (fix impact analysis) > Builder (implementation) Ripple > Guardian chain: Ripple (impact analysis) > Guardian (PR strategy) Field > Trace > Echo chain: Field (persona definition) > Trace (real-data validation) > Echo (simulation confirmation) Sentinel > Canon > Builder chain: Sentinel (vulnerability detection) > Canon (OWASP compliance evaluation) > Builder (fix implementation) Gateway > Canon > Gateway chain: Gateway (API design) > Canon (OpenAPI/RFC compliance check) > Gateway (corrections) Echo > Canon > Palette chain: Echo (UX issues) > Canon (WCAG compliance evaluation) > Palette (accessibility fixes) Field > Cast > Echo chain: Field (research data) > Cast (persona integration) > Echo (UI validation) Trace > Cast chain: Trace (behavioral data) > Cast (persona evolution)

Git/PR Management

Agent Description Output
Guardian "Every commit tells a story. Make it worth reading." - Git/PR gatekeeper. Signal/Noise analysis of changes, commit granularity optimization, branch naming, PR strategy proposals Analysis report, PR preparation
Launch "Shipping is not the end. It's the beginning of accountability." - Release management plus read-only GitHub PR collection, weekly/monthly/client reporting, DORA/SPACE metrics, retrospectives, and PDF export Release plans, CHANGELOG and delivery reports
Trail "Every bug has a birthday. Every regression has a parent commit. Find them." - Git history investigation, regression root cause analysis, code archaeology. Travels back in time to uncover the truth History investigation report

Guardian > Judge > Zen chain: Guardian (PR preparation) > Judge (review) > Zen (fixes) Guardian > Launch chain: Guardian (change analysis) > Launch (release plan) Trail > Scout chain: Trail (regression identification) > Scout (detailed investigation)

Quality Assurance

Agent Description Output
Radar "Untested code is unfinished code." - Unit/integration test addition, flaky test fixing, coverage improvement, and test-data factories/fixtures (fixtures Recipe) Test code
Voyager "E2E tests are the user's advocate in CI/CD." - Cross-platform and iOS E2E specialist. Playwright/Cypress/Appium/Detox/Maestro plus XCUITest and App Store snapshots E2E test code
Sentinel "Security is not a feature. It's a responsibility." - Static security analysis (SAST), vulnerability pattern detection, input validation Security fixes
Probe "A system is only as secure as its weakest endpoint." - Dynamic security testing (DAST), OWASP ZAP/Nuclei integration, penetration testing Vulnerability report
Vigil "An undetected attack is an undefended system. Vigil ensures nothing passes unseen." - Detection Engineering agent. Sigma/YARA rule design, detection coverage mapping (MITRE ATT&CK), threat hunting hypothesis design, Purple Team Blue side execution, Detection-as-Code CI/CD integration Detection rules, coverage maps
Judge "Good code needs no defense. Bad code has no excuse." - Code review via codex review, automated PR review, pre-commit checks, AI hallucination detection Review report
Zen "Clean code is not written. It's rewritten." - Refactoring and code quality improvement (behavior unchanged) Code improvements
Sweep "Dead code is technical debt that earns no interest." - Unused file detection, dead code identification, orphaned file discovery, safe deletion proposals Cleanup proposals
Attest "Specs are truth. Code is evidence. Attest finds the gaps." - Specification compliance verifier. Extracts acceptance criteria from specs, generates BDD scenarios, and adversarially probes for gaps between spec and implementation. Issues CERTIFIED/CONDITIONAL/REJECTED verdicts Compliance report, BDD scenarios
Siege "Break it before users do. Fix it before they notice." - Advanced testing specialist. Load testing (k6/Locust/Artillery), contract testing (Pact CDC), chaos engineering, mutation testing, resilience pattern verification Test results, resilience reports
Void "The best code is the code that was never written." - YAGNI enforcement, scope cutting, complexity reduction proposals. Challenges existence of every feature/abstraction with 5 questions and Cost-of-Keeping Score Subtraction proposals
Breach "Think like an attacker. Defend like an engineer." - Red team engineering. Attack scenario design, threat modeling, MITRE ATT&CK/OWASP frameworks, Purple Team exercises, AI/LLM red teaming Security assessment
Cloak "Privacy is not a feature. It's a right." - Privacy engineering and data governance. PII detection, data flow mapping, consent management, GDPR/CCPA-compliant code implementation Privacy assessment
Chain "Treat every third-party skill like an npm install. Audit before invoking." - Skill/plugin/MCP intake audit plus live npm/PyPI malware investigation. Generates sha256 manifests, scans Unicode and exfiltration patterns, matches campaign IoCs, and produces persistence-first eradication and credential-rotation runbooks Supply-chain audit, infection report and recovery runbook

Implementation

Agent Description Output
Builder "Types are contracts. Code is a promise." - Type-safe production implementation plus Gemini image-generation/editing code, prompt design, batch generation, style transfer, post-processing, provenance, and content-policy gates, plus CLI/TUI construction (cli) and regex/parser/DSL authoring (grammar) Production and image-generation code
Artisan "Prototypes promise. Production delivers." - Production frontend implementation craftsman. React/Vue/Svelte, Hooks design, state management, Server Components, form handling, data fetching Frontend code
Forge "Done is better than perfect. Ship it, learn, iterate." - Prototyping. Prioritizes working software over perfection. Outputs types.ts, errors.ts, forge-insights.md for Builder handoff MVP/PoC
Native "Every pixel ships. Every platform matters." - Pure-native mobile implementation specialist for iOS (Swift 6.3 + SwiftUI + Liquid Glass) and Android (Kotlin 2.4+ + Jetpack Compose + Material 3 Expressive). Production-quality features with @Observable / Swift Concurrency, Compose Strong Skipping + Type-safe Navigation, SwiftData / Room, Credential Manager + Passkeys, Privacy Manifest, edge-to-edge, predictive back, Live Activities, App Intents, Foundation Models / Gemini Nano, store compliance, and per-store staged rollout. React Native / Flutter / KMP / CMP are out of scope Code
Pixel "Every pixel matters. Fidelity is non-negotiable." - Faithful reproduction agent. Generates pixel-accurate HTML/CSS from image mockups (PNG/JPG/screenshots) and performs visual verification HTML/CSS code

AI/ML

Agent Description Output
Oracle "AI is only as good as its architecture. Design it, measure it, trust nothing." - AI/ML design and evaluation specialist. Prompt engineering, RAG architecture, LLM application patterns, safety guardrails, evaluation frameworks, MLOps, cost optimization Design specs, evaluation reports
Chisel "A vague word is a decision you left to chance. Carve it into something that can be checked." - Prompt-to-executable-specification translator. Detects vague quality/quantity/style/design/judgment wording, dissolves persona and title lines into capabilities, reconciles contradictory instructions, and records what was deliberately left open Ambiguity ledger, specified prompt

Oracle > Builder > Radar chain: Oracle (AI/ML design) > Builder (implementation) > Radar (tests) Oracle > Stream > Builder chain: Oracle (RAG design) > Stream (data pipeline) > Builder (implementation) Oracle > Sentinel > Oracle chain: Oracle (safety design) > Sentinel (security review) > Oracle (refinement)

Performance

Agent Description Output
Bolt "Speed is a feature. Slowness is a bug you haven't fixed yet." - Application performance improvement. Frontend (re-render reduction) and backend (N+1 fix) optimization Optimized code
Tuner "A fast query is a happy user. A slow query is a lost customer." - DB performance optimization. EXPLAIN ANALYZE analysis, index recommendations, slow query improvement Query optimization

Observability/SRE

Agent Description Output
Beacon "You can't fix what you can't see. You can't see what you don't measure." - Observability and reliability engineering specialist. SLO/SLI design, distributed tracing, alerting strategy, dashboard design, capacity planning, toil automation SLO definitions, observability specs

Beacon > Gear > Builder chain: Beacon (observability design) > Gear (monitoring implementation) > Builder (instrumentation) Triage > Beacon > Gear chain: Triage (incident postmortem) > Beacon (monitoring improvements) > Gear (implementation)

UI/UX

Agent Description Output
Vision "Design is not how it looks. Design is how it feels." - Creative direction. Design direction decisions, Design System creation, Muse/Palette/Flow/Forge orchestration Design strategy
Palette "Usability is invisible when done right, painful when done wrong." - Usability improvement, cognitive load reduction, a11y support UX improvements
Muse "Tokens are the DNA of design. Mutate them with care." - Design token application, spacing/border-radius/shadow unification, dark mode support Visual improvements
Flow "Motion creates emotion. Animation breathes life." - UI animation, hover effects, loading states, modal transitions Animations
Echo "I don't test interfaces. I feel what users feel." - Persona validation and synthetic demand. Reports UI confusion, feature requests, JTBD, and unmet needs UX report
Vitrine "Components without stories are components without context." - Storybook story creation, catalog management, Visual Regression integration. CSF 3.0 format Storybook Stories
Prose "Words are the smallest unit of design. Get them wrong, and nothing else matters." - User-facing text specialist. Microcopy, error messages, voice & tone framework, onboarding copy, accessibility text Copy guidelines, content specs
Frame "Design speaks in pixels. I translate it to code." - Figma MCP Server bridge agent. Extracts and structures design context from Figma for implementation agents. Design-to-code bridging, Code Connect management, design system rule extraction. No code written Structured design context, design system rules
Ink "Every stroke serves a purpose." - SVG icon/illustration generation, icon system design, and sprite symbol construction SVG assets

Documentation

Agent Description Output
Scribe "A specification is a contract between vision and reality." - Specification author for PRD/SRS/HLD/LLD and cross-team L0-L3 packages, including implementation checklists, test specifications, and document format conversion (convert Recipe) Specs, design docs
Quill "Code tells computers what to do. Documentation tells humans why." - JSDoc/TSDoc additions, README updates, typing any to proper type definitions Documentation
Tome "Changes are forgotten. Knowledge endures." - Transforms changes into learning documentation and verified knowledge into public technical articles for note/Zenn/Qiita/dev.to, including article series and repurposing Learning docs and technical articles

Scribe vs Quill vs Tome responsibilities:

  • Scribe: Project documentation (PRD, SRS, design docs, checklists, test specifications) and format conversion (Markdown > PDF/Word/HTML, convert Recipe)
  • Quill: Code documentation (JSDoc/TSDoc, README, type definitions)
  • Tome: Change-based learning materials and external-facing articles for note/Zenn/Qiita/dev.to, including series management

Visualization

Agent Description Output
Canvas "A diagram is worth a thousand lines of documentation." - Design visualization. Converts code, specs, and context into Mermaid diagrams or ASCII art (flowcharts, sequence diagrams, state machines, class diagrams, ER diagrams, etc.) Mermaid diagrams / ASCII Art
Cue "Every frame tells a story." - Video scripts, storyboards, narration, and Playwright feature-demo production with multi-aspect recording, captions, quality gates, voiceover, transcript, and thumbnails Video plans, demos and publication assets
Stage "Every slide is a stage." - Marp/reveal.js/Slidev slide generation, story composition design, conference talk optimization Slide decks

Architecture

Agent Description Output
Atlas "Dependencies are destiny. Map them before they map you." - Dependency analysis, circular reference detection, ADR/RFC creation Design documents
Port "From web to native. Translate the experience, not just the code." - Web-to-native porting design specialist (2026 spec — Liquid Glass / Material 3 Expressive / Swift 6.3 / targetSdk 36 / Privacy Manifest / 5.1.2(i) AI disclosure aware). Designs porting blueprints from Web (React/Vue/Svelte/Angular) to iOS Swift / Android Kotlin pure-native. Produces feature parity matrices, native architecture maps, regulatory-compliance plans, and Strangler-Fig phased roadmaps. Optionally proposes a hybrid path (Pure-Native UI + KMP shared logic) Porting blueprint, parity matrix, roadmap
Gateway "APIs are promises to the future. Design them like contracts." - API design, review, OpenAPI spec generation, versioning strategy, breaking change detection, and messaging-platform integration (messaging Recipe) API specifications
Grove "A well-structured repository is a well-structured mind." - Human- and LLM-optimized repository structure design, including context efficiency, prompt-cache topology, and sharding Structure design, audit reports
Weave "Every state tells a story. Every transition is a contract." - Workflow and state machine design specialist. State transition design, invalid transition detection, Saga patterns, approval flows, and cron/timezone/business-calendar scheduling (schedule Recipe) Designs, diagrams
Seek "The right result at the right time in the right order." - Search engine and vector DB design specialist. Full-text search, vector search, hybrid search, RAG retrieval layer Code, configs
Crypt "Trust no channel. Verify every key." - Cryptographic architecture design: algorithm selection, key management, E2E encryption, KMS integration, TLS configuration Crypto design specs

Communication

Agent Description Output

Gateway > Builder > Radar chain: Gateway (messaging design) > Builder (implementation) > Radar (tests) Gateway > Gateway chain: Gateway (webhook API spec) > Gateway (handler design)

Data

Agent Description Output
Schema "A schema is a contract with the future." - DB schema, migration, ER diagram, and multi-tenant isolation/RLS/routing design Migrations / schema definitions
Stream "Data flows like water. My job is to build the pipes." - Data pipelines. ETL/ELT design, Kafka/Airflow/dbt, batch/streaming selection, data quality management Pipeline design, DAGs, dbt models

Schema > Stream chain: Schema (data model) > Stream (pipeline design)

DevOps

Agent Description Output
Gear "The best CI/CD is the one nobody thinks about." - Dependency management, CI/CD and Docker optimization, plus advanced GitHub Actions workflow design Configuration files
Scaffold "Infrastructure is the silent foundation of every dream." - Cloud infrastructure (Terraform/CloudFormation/Pulumi), local dev environments (Docker Compose), IaC design Infrastructure config
Hone "A sharp blade cuts clean. A sharp config cuts friction." - AI CLI configuration auditor and Claude Code lifecycle-hook specialist. Audits Codex/Antigravity/Claude configuration and designs, configures, or debugs scoped hooks with verification. Also owns the personal environment — dotfiles, shell/editor config, and macOS automation (env / automate Recipes) Audit reports, proposals and hook configurations
Ledger "Every dollar has a story. Make it a short one." - FinOps and cloud cost optimization. IaC cost estimation, right-sizing, RI/SP recommendations, cost anomaly detection Reports, configs
Shift "Migration is not moving. It's transforming." - Migration and upgrade orchestrator. Framework, library, API, database, and infrastructure migrations end-to-end with codemod generation and incremental strategies Migration plans

Builder vs Hone vs Gear vs Scaffold responsibilities:

  • Builder: CLI and TUI construction (cli Recipe) — the tool itself
  • Hone: Personal environment (dotfiles, shell, editor, terminal, macOS automation) plus AI CLI configuration audit and Claude Code event-hook design/configuration/debugging
  • Gear: Project-level DevOps (CI/CD, Docker, monitoring, Git hooks); gha mode owns advanced GitHub Actions
  • Scaffold: Infrastructure provisioning (cloud, Docker Compose, IaC)

Internationalization

Agent Description Output
Polyglot "Every language deserves respect. Every user deserves their mother tongue." - i18n support. Replaces hardcoded strings with t() functions, formats dates/currencies with Intl API i18n implementation

Growth

Agent Description Output
Growth "Traffic without conversion is just expensive vanity." - SEO (meta/OGP/JSON-LD), SMO (social share display), CRO (CTA improvement), and retention/re-engagement design (retention Recipe) Growth initiatives
Funnel "Above the fold is your one shot. Make every pixel convert." - LP (Landing Page) conversion strategist and premium production orchestrator. Framework-based structure design (AIDA/PAS/BAB/4Ps), hero section craft, CTA placement strategy, social proof hierarchy, mobile-first responsive implementation LP structure, copy, specs

Analytics

Agent Description Output
Pulse "What gets measured gets managed. What gets measured wrong gets destroyed." - KPI definition, tracking event design, dashboard spec creation Metrics design
Experiment "Every hypothesis deserves a fair trial. Every decision deserves data." - A/B test design, hypothesis documentation, sample size calculation, feature flag implementation Experiment reports

Operations

Agent Description Output
Triage "In chaos, clarity is the first act of healing." - Incident response, impact assessment, recovery procedures, postmortem creation Operations reports
Mend "Known failures deserve known fixes. Speed of recovery defines reliability." - Auto-repair agent for known failure patterns. Receives Triage diagnostics and Beacon alerts, executes runbooks based on safety tier classification, performs staged verification, and manages rollback Repair results, verification reports

Triage > Mend > Beacon chain: Triage (diagnosis) > Mend (auto-repair) > Beacon (monitoring recovery)

Browser Automation

Agent Description Output
Vector "The browser is a stage. Every click is a scene." - Browser automation via Playwright/Chrome DevTools. Data collection, form interaction, screenshots, network monitoring, and crawl-system architecture (crawl Recipe) Automation scripts

Cue vs Vitrine responsibilities:

  • Cue: Browser (Web UI) demo videos (Playwright, .webm output)
  • Vitrine: Component catalogs and Storybook stories with visual-regression wiring

Workflows

Basic Usage

  1. Invoke an agent with /AgentName
  2. The agent executes the task
  3. Suggests handoff to other agents as needed

Orchestration with Nexus

Use Nexus for complex tasks. Nexus operates in the following modes:

Mode Trigger Behavior Interaction
Full Auto ## NEXUS_AUTORUN + simple task Fully automatic execution Only on errors
Guided ## NEXUS_GUIDED or default Confirms at decision points Option-based interaction
Interactive ## NEXUS_INTERACTIVE Confirms at each step Always interactive
Continue ## NEXUS_HANDOFF Result handoff Interaction as needed

Interactive Execution (Guided/Interactive)

Each agent asks for user confirmation at important decision points (using the platform's interaction capabilities):

  • Start confirmation: Confirms approach after chain design, before execution
  • Decision point confirmation: Security risks, destructive changes, multiple approaches, etc.
  • Questions as options: Choose from 2-4 options ("Other" is always available)
# Example option-based question
questions:
  - question: "A potential security vulnerability was found. How would you like to proceed?"
    header: "Security"
    options:
      - label: "Audit with Sentinel (Recommended)"
        description: "Request review from the security specialist agent"
      - label: "Continue with acknowledged risk"
        description: "Proceed at your own risk"
      - label: "Abort investigation"
        description: "Stop for safety"

Automatic Mode Selection by Complexity

Indicator SIMPLE COMPLEX
Estimated steps 1-2 3+
Affected files 1-3 4+
Security-related No Yes
Destructive changes No Yes
  • SIMPLE + NEXUS_AUTORUN: Fully automatic execution
  • COMPLEX: Automatically switches to Guided mode (interaction required)

See _common/INTERACTION.md for details.

Chain Templates by Task Type

Investigation & Understanding

Task Description Chain
INVESTIGATE/feature Feature existence/implementation investigation Lens
INVESTIGATE/flow Data flow/processing flow tracing Lens > Canvas
INVESTIGATE/onboarding Full codebase understanding Lens > Scribe
INVESTIGATE/pre-impl Investigation then implementation Lens > Builder > Radar

Lens vs Scout: Lens = codebase understanding and feature exploration ("Does X exist?", "How does Y flow?"), Scout = bug investigation and root cause analysis ("Why did it break?")

Bug Fixes

Task Description Chain
BUG/simple Simple bug fix Scout > Builder > Radar
BUG/complex Complex bug (RCA required) Scout > Sherpa > Builder > Radar > Sentinel
BUG/frontend Frontend bug Scout > Artisan > Radar

Feature Development

Task Description Chain
FEATURE/S Small feature Builder > Radar
FEATURE/M Medium feature Sherpa > Forge > Builder > Radar
FEATURE/L Large feature Spark > Sherpa > Forge > Builder > Radar > Quill
FEATURE/frontend Frontend feature Sherpa > Forge > Artisan > Radar
FEATURE/fullstack Full-stack feature Sherpa > Forge > Artisan > Builder > Radar
FEATURE/api API development Gateway > Builder > Radar

UI/UX

Task Description Chain
UI/new New UI implementation Vision > Forge > Vitrine > Muse > Artisan > Radar
UI/redesign UI redesign Vision > Muse > Palette > Flow > Artisan > Radar
UI/component Component creation Forge > Vitrine > Muse > Artisan > Radar
UI/animation Animation addition Flow > Artisan > Radar
UX/research UX research Field > Echo > Palette
UX/improve UX improvement Echo > Palette > Artisan > Radar
UX/session-analysis Session analysis Trace > Echo > Palette
UX/persona-validation Persona validation Field > Trace > Echo

Refactoring

Task Description Chain
REFACTOR/small Small refactor Zen > Radar
REFACTOR/arch Architecture improvement Atlas > Sherpa > Zen > Radar
REFACTOR/legacy Legacy modernization Shift (detect) > Sherpa > Zen > Radar

Performance

Task Description Chain
PERF/frontend Frontend optimization Bolt > Artisan > Radar
PERF/backend Backend optimization Bolt > Builder > Radar
PERF/db Database optimization Tuner > Schema > Builder > Radar

AI/ML

Task Description Chain
AI/rag RAG pipeline design Oracle > Stream > Builder > Radar
AI/llm-app LLM application design Oracle > Builder > Radar
AI/safety AI safety review Oracle > Sentinel > Oracle
AI/prompt-ops Prompt engineering & evaluation Oracle > Radar
AI/prompt-spec Turning a vague prompt into an executable spec Chisel

Observability/SRE

Task Description Chain
SRE/slo SLO definition & monitoring Beacon > Gear > Builder
SRE/observability Full observability setup Beacon > Gear > Builder > Radar

Security

Task Description Chain
SECURITY/audit Static analysis Sentinel > Builder > Radar
SECURITY/pentest Dynamic testing Probe > Builder > Radar > Probe
SECURITY/full Full audit Sentinel > Probe > Builder > Radar > Sentinel

Testing

Task Description Chain
TEST/unit Unit test addition Radar
TEST/e2e E2E test addition Voyager
TEST/coverage Coverage improvement Radar > Voyager
TEST/load Load testing Siege > Bolt
TEST/chaos Chaos engineering Siege > Triage > Builder
TEST/contract Contract testing Gateway > Siege > Radar
TEST/mutation Mutation testing Siege > Radar

Review

Task Description Chain
REVIEW/pr PR review Judge > Zen/Builder/Sentinel
REVIEW/security Security review Judge > Sentinel

Git/PR

Task Description Chain
GIT/pr-prep PR preparation Guardian > Judge
GIT/commit-split Commit splitting Guardian
GIT/pr-full Implement > PR > Review Builder > Guardian > Judge > Zen
GIT/release Release notes generation Guardian

Decision Making

Task Description Chain
DECISION/arch Architecture selection Magi > Builder/Zen
DECISION/strategy Strategic decisions Magi > Spark
DECISION/office-hours Founder office hours advisory Magi > Builder/Echo[demand]/Sherpa
DECISION/retro Postmortem on a recent decision/outcome Magi

Analysis

Task Description Chain
ANALYSIS/impact Change impact analysis Ripple > Builder > Radar
ANALYSIS/standards Standards compliance check Canon > Builder > Radar
ANALYSIS/cleanup Code cleanup Sweep > Zen > Radar

Subtraction & Culture

Task Description Chain
SUBTRACT/feature-gate Feature proposal subtraction gate Spark > Void > Magi
SUBTRACT/scope-check Scope validation Sherpa > Void > Sherpa
SUBTRACT/arch-simplify Architecture over-engineering detection Atlas > Void > Zen