claude

ainous-team

Persistent agent team with evolving playbooks, trust progression, and authority governance

1 on GitHubv5.21.1updated 4 months ago
/

Plugin ainous-team from marketplace ainous-team in xAlbert1d/Ainous-Team. Commands read from the repo's manifest — 1 month ago.

This plugin runs code on your machine — its hooks fire automatically on events like tool use. The exact commands are listed below; read them before installing.

What it installs

Skills (63)

  • a11y

    Accessibility compliance — WCAG, screen readers, keyboard navigation, color contrast, semantic HTML, ARIA. Use when building or reviewing UI, ensuring legal compliance, or improving usability for all users.

  • api-design

    API design principles and contracts. Use when designing, reviewing, or modifying HTTP/REST APIs. Invoke when defining endpoints, error formats, pagination, or versioning strategy.

  • auto-decide

    Decision classification framework — separates mechanical decisions (auto-resolve) from taste decisions (need human input). Use when triaging multiple decisions, planning workflows, or reducing decision fatigue.

  • caption-format

    Formats captions and subtitles for readability, timing, and accessibility. Produces SRT/VTT format output. Use when publishing videos, ensuring accessibility compliance, or adding multilingual subtitles.

  • code-review-ext

    Comprehensive multi-pass code review covering architecture, logic, performance, security, and conventions. Use for pull request reviews, pre-merge checks, and code audits. Extends the verify skill with proactive quality analysis.

  • codex-image-gen

    Generate production raster images by driving the OpenAI Codex CLI (gpt-image-2). Use when you need real image assets — hero art, textures, icons, thumbnails, OG/social cards — created and saved into a project without your own image-API key. Triggers: "generate an image", "make a hero/background/icon

  • competitive-intel

    Structured comparison analysis of tools, frameworks, or approaches. Use for technology selection, architecture decisions, tool migration evaluation, or market analysis.

  • compliance-check

    Regulatory compliance review for code and data systems. Use when code handles PII, PHI, payment data, or operates in regulated environments (GDPR, HIPAA, SOC 2, PCI-DSS). Invoke before shipping features that touch user data collection, retention, or deletion.

  • confidence-calibration

    Structural confidence labeling for role outputs. Use when emitting findings, assessments, or recommendations that downstream roles or the coordinator will act on. Invoke before any completed event where genuine uncertainty exists.

  • content-repurpose

    Converts long-form content into multiple formats while preserving core message. Use when turning design docs into READMEs, research into summaries, or internal docs into external-facing content.

  • contract-testing

    Service contract verification. Use when testing interactions between services — HTTP APIs, message queues, event streams. Invoke when integration tests are too slow or brittle, or when you need to verify that producer and consumer agree on the data contract without deploying both.

  • copywriting

    Generates clear, impactful text with hooks and structured flow. Use when writing READMEs, release notes, product descriptions, project descriptions, or blog posts.

  • data-model

    Schema design, normalization, denormalization tradeoffs, indexing strategy, and data lifecycle. Use when designing databases, planning migrations, or evaluating data architecture decisions.

  • debug

    Systematic debugging methodology. Use when fixing bugs, investigating test failures, or diagnosing unexpected behavior. Invoke before proposing any fix.

  • deep-research

    Synthesizes insights from multiple sources — filters noise, identifies patterns, produces actionable summaries. Use for technology evaluation, paper analysis, codebase archaeology, or market research.

  • design

    Architecture and design principles. Use when designing features, planning implementations, or producing design artifacts for handoff. Invoke when the task requires structural decisions.

  • devops

    Guides deployment, CI/CD, infrastructure, and automation tasks with concrete commands and configuration. Use for CI/CD setup, deployment troubleshooting, infrastructure changes, and release management.

  • diagram

    Converts concepts, architectures, and workflows into renderable diagrams using Mermaid syntax. Use when creating architecture docs, design handoffs, README visuals, or system overviews.

  • docs

    Technical documentation principles. Use when writing READMEs, guides, API docs, inline comments, or any prose that another human must act on. Invoke before drafting any documentation artifact.

  • estimate

    Effort estimation, complexity assessment, and uncertainty communication. Use when scoping work, planning timelines, or communicating delivery expectations. Covers point estimates, ranges, and risk factors.

  • flowchart

    Generates decision trees and process flowcharts from textual descriptions using Mermaid syntax. Use for troubleshooting guides, architecture decisions, deployment processes, and user flows.

  • ideate

    Structured product ideation and creative exploration. Use when exploring what to build, evaluating opportunities, brainstorming features, or validating demand before committing to implementation.

  • image-background

    Author a gpt-image-2 full-bleed abstract/ambient background prompt and hand off to codex-image-gen. Triggers: background art, abstract backdrop, ambient gradient art.

  • image-craft-base

    Shared craft conventions for the image-* gpt-image-2 skills — 5-section prompt template, style-anchor format, size/aspect table, and the transparency/text/tiling/safe-zone rules. Reference note for the image-* skills; not invoked directly.

  • image-hero

    Author a gpt-image-2 hero/banner prompt and hand off to codex-image-gen. Triggers: make a hero, banner image, landing page background art.

  • image-icon

    Author a gpt-image-2 prompt for an app icon, favicon, glyph, or logo/wordmark and hand off to codex-image-gen. Triggers: make an icon, app icon/favicon, logo/glyph, transparent icon.

  • image-illustration

    Author a gpt-image-2 spot-illustration or CONSISTENT-series prompt and hand off to codex-image-gen. Triggers: spot illustration, illustration set, matching series, consistent illustrations.

  • image-social-card

    Author a gpt-image-2 OG/social card or 9:16 story prompt with verbatim text and hand off to codex-image-gen. Triggers: OG image, social card, Open Graph, share card, story graphic.

  • image-texture

    Author a gpt-image-2 SEAMLESS tileable texture prompt and hand off to codex-image-gen. Triggers: tileable texture, seamless pattern, repeating background pattern.

  • image-thumbnail

    Author a gpt-image-2 thumbnail prompt and hand off to codex-image-gen. Triggers: make a thumbnail, video thumbnail, card image, list thumbnail.

  • impact-analysis

    Change impact assessment. Use before modifying any shared system, API contract, or widely-used component. Invoke when the blast radius of a change is unclear. Skip for isolated, single-file changes with no consumers.

  • infographic

    Turns data and processes into structured visual summaries using markdown tables, metrics, and hierarchy. Use for project reports, dashboards, status summaries, and team presentations.

  • knowledge-structure

    Organizes unstructured information into clear frameworks with hierarchy and cross-references. Use when structuring team-knowledge, organizing research output, or designing documentation architecture.

  • migrate

    Migration and upgrade methodology. Use when planning schema changes, data migrations, API version upgrades, or any operation that transforms existing production state. Invoke before executing any migration.

  • negotiate

    Managing conflicting requirements, stakeholder alignment, technical vs business tradeoffs, and saying no constructively. Use when facing competing demands, scope conflicts, or resource contention.

  • observability

    Logging, metrics, tracing, and alerting strategy. Use when designing monitoring for a service, investigating production behavior, or setting up alerting. Covers the three pillars and runbook design.

  • onboard

    How to ramp up on a new codebase, project, or team. Systematic codebase archaeology, finding entry points, understanding conventions. Use when joining a project, starting a new engagement, or helping someone get oriented.

  • perf

    Performance optimization principles. Use when investigating slow code, optimizing hot paths, adding caching, or evaluating performance tradeoffs. Invoke before making any change motivated by speed.

  • post-mortem

    Structured incident analysis after production failures. Use after outages, data loss, security breaches, or any event that impacted users. Blameless methodology with timeline, root cause, and prevention.

  • premise-check

    Strategic premise validation before committing to a solution. Use before any significant implementation, product decision, architecture change, or resource allocation. Challenges whether the problem is correctly framed.

  • present

    Technical presentation to any audience — storytelling with data, slide structure, demo scripting, adapting depth to audience. Use when preparing talks, demos, stakeholder updates, or any presentation of technical work.

  • prioritize

    Frameworks for ranking competing opportunities — RICE, ICE, MoSCoW, and weighted scoring. Use when triaging backlogs, planning roadmaps, or deciding what to work on next.

  • refactor

    Refactoring methodology and discipline. Use when improving code structure, reducing complexity, or cleaning up before adding features to tangled code.

  • release-gate

    Release engineering discipline — scope drift detection, coverage audit, test bootstrap, and documentation sync. Use before shipping any release, creating a PR, or deploying to production.

  • retro-metrics

    Quantified retrospective with shipping metrics, health trends, and growth tracking. Use for periodic team reviews, sprint retrospectives, or any reflection that should be data-driven rather than feelings-driven.

  • review-response

    How to evaluate and respond to code review feedback. Use when receiving review comments, PR feedback, or quality findings from another role. Invoke before implementing any review suggestion.

  • runbook-creation

    Operational runbook authoring. Use when creating or updating incident response procedures, deployment steps, rollback guides, or on-call playbooks. Invoke when a process must be reproducible by someone who wasn't present when it was designed.

  • scqa

    Structures content using the Situation→Complication→Question→Answer framework for clear logical narratives. Use when writing architecture proposals, RFCs, incident reports, or design documents.

  • security-scan

    Security scanning methodology. Use when reviewing code for vulnerabilities, auditing dependencies, or checking for secrets and misconfigurations before release.

  • skill-creator

    Meta-skill for creating new AI skills in .md format. Use when designing, structuring, or improving skills for the vault.

  • source-validate

    Validates credibility and reliability of information sources. Use when processing external signals, evaluating technology claims, or assessing research findings.

  • strategy-evolution

    Playbook evolution methodology for the consolidator role. Use when adding, modifying, or retiring playbook strategies. Invoke during consolidation cycles.

  • structural-isolation

    Multi-agent isolation methodology. Use when spawning parallel reviewers, running map-reduce analysis, or performing perspective forking — any task where independent outputs are required before aggregation.

  • summarize

    Condenses complex content into actionable summaries. Use for journal compaction, compiled truth synthesis, research distillation, and any long-form-to-short-form transformation.

  • tdd

    Test-driven development principles. Use when implementing features or fixing bugs where tests exist or should exist. Invoke when you're about to write code that should be tested.

  • test-strategy

    Deciding which types of tests to use where — unit, integration, e2e, load, contract, chaos. Broader than TDD (which is a write-test-first methodology). Use when designing a testing approach for a project or feature.

  • threat-model

    Structured threat modeling using STRIDE framework and trust boundary mapping. Use when designing security-sensitive features, reviewing architecture for threats, or preparing for security audits. Extends security-scan with architecture-level thinking.

  • tone-enforce

    Ensures consistent voice and style across outputs — code comments, docs, commit messages, error messages. Use when reviewing multi-author docs or maintaining project voice.

  • ui-layout
  • verify
  • video-edit
  • video-script
  • workflow-auto

Slash commands (9)

  • /ainous-team:team-alerts

    Show health alerts for team roles

  • /ainous-team:team-history

    Show session history for all team roles

  • /ainous-team:team-implement

    Run a full team implementation pipeline — research, design, code, test, and review for end-to-end feature development.

  • /ainous-team:team-init

    Initialize Ainous Team for the first time. Guides you through choosing coordinator mode and sets up all 13 roles.

  • /ainous-team:team-retro

    Run a periodic team health review — individual growth, team dynamics, coordinator self-assessment, and action items.

  • /ainous-team:team-review-periodic

    Periodic team health review — individual growth, team dynamics, coordinator self-assessment, and action items. Triggered by /team-retro or automatically every 7 days / 10 commits.

  • /ainous-team:team-review

    Run a full team review pipeline on code or architecture — security, quality, and architecture in parallel.

  • /ainous-team:team-signal

    Scan external sources for team-relevant signals — trends, tools, vulnerabilities, and insights from GitHub, HackerNews, social platforms, RSS feeds, and more.

  • /ainous-team:team-status

    Show the persistent roles team dashboard

Subagents (13)

  • architect

    Designs system architecture, evaluates trade-offs. Refuses to accept the current problem framing, documents rejected alternatives, kills designs that evidence disproves. <example>Context: User needs architecture design\nuser: \"design the auth system\"\nassistant: \"I'll use the architect agent.\"\n

  • authority

    Permission gating, policy enforcement, escalation decisions. Reasons from precedent, never hedges approvals, watches for patterns that signal baseline expansion.

  • code-quality

    Reviews code for bugs, style, maintainability. Distinguishes wrong from merely different, always includes the better approach, ends reviews with a verdict.

  • consolidator

    Distills team learnings, evolves playbooks, maintains knowledge base. Trusts traces over summaries, conservative about promoting patterns, questions aging strategies.

  • coordinator

    Orchestrates the team, routes tasks, synthesizes results. Challenges briefs that sound too clean before spawning. Never implements directly. <example>Context: User wants to implement a feature end-to-end\nuser: \"@coordinator implement user auth for the API\"\nassistant: \"I'll use the coordinator t

  • designer

    Owns brand identity, UX flows, UI specifications, and design systems. Advocates for the user and the brand; resists generic AI aesthetics and cookie-cutter layouts. Produces structured design specs — not finished assets — so developers can implement with confidence. <example> User: "The onboarding f

  • developer

    Implements features, fixes bugs, writes production code. Reads tests before writing, never claims done without verification, resists scope creep. <example>Context: User wants code implemented\nuser: \"implement the login endpoint\"\nassistant: \"I'll use the developer agent to write the implementati

  • researcher

    Explores codebase, investigates technologies, produces findings. Does not trust first answers, labels confidence explicitly, surfaces constraints nobody asked about.

  • retriever

    Filters role knowledge by task relevance. Matches by meaning not keywords, returns sparse high-signal context, says 'nothing relevant here' plainly when that's the truth.

  • security

    Security reviews, threat modeling, vulnerability scanning. Assumes adversarial input, never softens findings, tests safety assumptions rather than accepting them.

  • signal

    External intelligence scanning — GitHub, HN, ArXiv, blogs. Surfaces signals that challenge assumptions, annotates relevance, enforces serendipity quota.

  • tester

    Writes tests, validates implementations. Assumes code is broken until proven otherwise, hunts hidden assumptions, tests the spec not just the code. <example>Context: User needs tests\nuser: \"write tests for the payment module\"\nassistant: \"I'll use the tester agent.\"\n<commentary>Testing task.</

  • writer

    Creates documentation, READMEs, changelogs. Leads with what readers need to do, verifies accuracy before publishing, writes in imperative second-person.

Hooks (10)

  • SessionEnd

    "${CLAUDE_PLUGIN_ROOT}/hooks/session-end"

  • SessionEnd

    "${CLAUDE_PLUGIN_ROOT}/hooks/teammate-lifecycle-reaper"

  • SubagentStop

    "${CLAUDE_PLUGIN_ROOT}/hooks/teammate-lifecycle-reaper"

  • SessionStart

    "${CLAUDE_PLUGIN_ROOT}/hooks/session-start"

  • PreToolUse

    "${CLAUDE_PLUGIN_ROOT}/hooks/authority-enforce.sh"

  • PostToolUse

    "${CLAUDE_PLUGIN_ROOT}/hooks/skill-telemetry"

  • PostToolUse

    "${CLAUDE_PLUGIN_ROOT}/hooks/spawn-telemetry"

  • PostToolUse

    "${CLAUDE_PLUGIN_ROOT}/hooks/webfetch-injection-detector"

  • PostToolUse

    "${CLAUDE_PLUGIN_ROOT}/hooks/taint-flag"

  • PostToolUse

    "${CLAUDE_PLUGIN_ROOT}/hooks/write-proxy"

View source on GitHub

Maintain ainous-team?

This page updates itself from your repo. Add the badge to your README, or get it Featured to pin it above the organic results with a followed link to your site.

claude & codex pluginslisted