code-ai-installer 4.0.0 → 4.0.1-b

package/domains/content/locales/en/.agents/skills/karpathy-guidelines/SKILL.md

@@ -1,6 +1,34 @@
 ---
 name: karpathy-guidelines
 description: Behavioral guidelines to reduce common LLM mistakes — think before coding, simplicity first, surgical changes, goal-driven execution.
+type: mandatory
+domain: content
+owners:
+  - conductor
+  - copywriter
+  - researcher
+  - reviewer
+  - strategist
+  - visual_concept
+gates:
+  - STRATEGIST
+  - RESEARCHER
+  - COPYWRITER
+  - VISUAL_CONCEPT
+  - REVIEWER
+  - RELEASE_GATE
+tech: []
+topic: []
+triggers:
+  - karpathy-guidelines
+  - karpathy
+  - Karpathy Guidelines
+  - сначала думай потом делай
+  - простота прежде всего
+  - хирургические изменения
+related: []
+budget_lines: 354
+schema_version: 1
 license: MIT
 source: https://x.com/karpathy/status/2015883857489522876
 ---
@@ -294,9 +322,10 @@ def test_sort_with_duplicate_scores():
         {'name': 'Charlie', 'score': 90},
     ]
     result = sort_scores(scores)
+    # Problem: with equal scores the order is unpredictable
     assert result[0]['score'] == 100
     assert result[2]['score'] == 90
-    assert result[0]['name'] < result[1]['name']  # stable sort
+    assert result[0]['name'] < result[1]['name']  # expect a stable sort
 
 # Verify: test fails → bug reproduced
 

package/domains/content/locales/en/.agents/skills/mailerlite-email-ops/SKILL.md

@@ -1,6 +1,21 @@
 ---
 name: mailerlite-email-ops
 description: Email management via MailerLite MCP — subscribers, segments, campaigns, automations, analytics. Direct operation via MCP tools.
+type: triggered
+domain: content
+owners:
+  - copywriter
+gates:
+  - COPYWRITER
+tech: []
+topic: []
+triggers:
+  - "mailerlite-email-ops"
+  - "Операции через MailerLite MCP"
+  - "Управление email через MailerLite MCP — подписчики, сегменты, кампании, автоматизации, аналитика. Прямая работа через MCP-tools"
+related: []
+budget_lines: 543
+schema_version: 1
 metadata:
   version: "2.0"
   mcp_server: "MailerLite MCP"

package/domains/content/locales/en/.agents/skills/marketing-psychology/SKILL.md

@@ -1,6 +1,21 @@
 ---
 name: marketing-psychology
 description: Behavioral science and mental models for marketing decisions — psychological triggers with scoring by leverage and feasibility
+type: triggered
+domain: content
+owners:
+  - copywriter
+gates:
+  - COPYWRITER
+tech: []
+topic: []
+triggers:
+  - "marketing-psychology"
+  - "Психология маркетинга"
+  - "Поведенческая наука и ментальные модели для маркетинговых решений — психологические триггеры с scoring по leverage и feasibility"
+related: []
+budget_lines: 414
+schema_version: 1
 ---
 # Marketing Psychology
 

package/domains/content/locales/en/.agents/skills/moodboard/SKILL.md

@@ -1,6 +1,21 @@
 ---
 name: moodboard
 description: Moodboard — visual concept tied to a brand, persona, and platform — references, palette, typography, Do's/Don'ts
+type: triggered
+domain: content
+owners:
+  - visual_concept
+gates:
+  - VISUAL_CONCEPT
+tech: []
+topic: []
+triggers:
+  - "moodboard"
+  - "Мудборд"
+  - "Мудборд — визуальная концепция с привязкой к бренду, персоне, платформе — референсы, палитра, типографика, Do/Don't"
+related: []
+budget_lines: 261
+schema_version: 1
 ---
 # Moodboard
 

package/domains/content/locales/en/.agents/skills/platform-compliance/SKILL.md

@@ -1,6 +1,21 @@
 ---
 name: platform-compliance
 description: Checking content compliance against platform rules — requirements registry, algorithmic risks, severity, and remediation recommendations
+type: triggered
+domain: content
+owners:
+  - reviewer
+gates:
+  - REVIEWER
+tech: []
+topic: []
+triggers:
+  - platform-compliance
+  - Соответствие платформам
+  - Проверка соответствия контента правилам платформ — реестр требований, алгоритмические риски, severity, рекомендации по исправлению
+related: []
+budget_lines: 324
+schema_version: 1
 ---
 # Platform Compliance
 

package/domains/content/locales/en/.agents/skills/platform-strategy/SKILL.md

@@ -1,6 +1,21 @@
 ---
 name: platform-strategy
 description: Platform Strategy — channel selection, funnel, paid/organic, resource estimation, growth roadmap, metrics with benchmarks
+type: triggered
+domain: content
+owners:
+  - copywriter
+gates:
+  - COPYWRITER
+tech: []
+topic: []
+triggers:
+  - "platform-strategy"
+  - "Стратегия платформ"
+  - "Стратегия по платформам — выбор каналов, воронка, paid/organic, оценка ресурсов, roadmap роста, метрики с бенчмарками"
+related: []
+budget_lines: 292
+schema_version: 1
 ---
 # Platform Strategy
 

package/domains/content/locales/en/.agents/skills/platform-visual-specs/SKILL.md

@@ -1,6 +1,21 @@
 ---
 name: platform-visual-specs
 description: Complete visual specifications guide — photo, video, ad formats, safe zones, accessibility, dark mode
+type: reference
+domain: content
+owners:
+  - visual_concept
+gates:
+  - VISUAL_CONCEPT
+tech: []
+topic: []
+triggers:
+  - "platform-visual-specs"
+  - "Спецификации визуалов"
+  - "Полный справочник визуальных спецификаций — фото, видео, рекламные форматы, safe zones, accessibility, dark mode"
+related: []
+budget_lines: 371
+schema_version: 1
 ---
 # Platform Visual Specs
 

package/domains/content/locales/en/.agents/skills/readability-scoring/SKILL.md

@@ -1,6 +1,21 @@
 ---
 name: readability-scoring
 description: Readability Scoring — adaptive norms per platform, readability indices, scanability, "fluff", suggested fixes
+type: triggered
+domain: content
+owners:
+  - reviewer
+gates:
+  - REVIEWER
+tech: []
+topic: []
+triggers:
+  - "readability-scoring"
+  - "Оценка читаемости"
+  - "Оценка читаемости — адаптивные нормы по платформам, индексы читаемости, scanability, «вода», suggested fixes"
+related: []
+budget_lines: 298
+schema_version: 1
 ---
 # Readability Scoring
 

package/domains/content/locales/en/.agents/skills/seo-copywriting/SKILL.md

@@ -1,6 +1,21 @@
 ---
 name: seo-copywriting
 description: SEO copywriting — SERP analysis, E-E-A-T, Featured Snippets, internal linking, schema markup, image optimization
+type: triggered
+domain: content
+owners:
+  - copywriter
+gates:
+  - COPYWRITER
+tech: []
+topic: []
+triggers:
+  - "seo-copywriting"
+  - "SEO-копирайтинг"
+  - "SEO-копирайтинг — SERP-анализ, E-E-A-T, Featured Snippets, internal linking, schema markup, оптимизация изображений"
+related: []
+budget_lines: 409
+schema_version: 1
 ---
 # SEO Copywriting
 

package/domains/content/locales/en/.agents/skills/social-media-formats/SKILL.md

@@ -1,6 +1,21 @@
 ---
 name: social-media-formats
 description: Social media formats — 7 platforms, format templates (carousel, thread, reel), engagement techniques, ToV adaptation, algorithmic best practices
+type: reference
+domain: content
+owners:
+  - copywriter
+gates:
+  - COPYWRITER
+tech: []
+topic: []
+triggers:
+  - "social-media-formats"
+  - "Форматы соцсетей"
+  - "Форматы соцсетей — 7 платформ, форматные шаблоны (карусель, thread, reel), engagement-приёмы, адаптация ToV, алгоритмические best practices"
+related: []
+budget_lines: 481
+schema_version: 1
 ---
 # Social Media Formats
 

package/domains/content/locales/en/.agents/skills/source-verification/SKILL.md

@@ -1,6 +1,21 @@
 ---
 name: source-verification
 description: Source verification — CRAAP framework, source types, bias checking, strictness levels, credibility rating
+type: triggered
+domain: content
+owners:
+  - reviewer
+gates:
+  - REVIEWER
+tech: []
+topic: []
+triggers:
+  - "source-verification"
+  - "Верификация источников"
+  - "Верификация источников — CRAAP-фреймворк, типы источников, bias-проверка, уровни строгости, рейтинг достоверности"
+related: []
+budget_lines: 284
+schema_version: 1
 ---
 # Source Verification
 

package/domains/content/locales/en/.agents/skills/storytelling-framework/SKILL.md

@@ -1,6 +1,21 @@
 ---
 name: storytelling-framework
 description: Storytelling frameworks — 10 frameworks, Decision Tree, emotional arc, platform adaptation, examples
+type: triggered
+domain: content
+owners:
+  - copywriter
+gates:
+  - COPYWRITER
+tech: []
+topic: []
+triggers:
+  - "storytelling-framework"
+  - "Фреймворк сторителлинга"
+  - "Фреймворки сторителлинга — 10 frameworks, Decision Tree, emotional arc, платформенная адаптация, примеры"
+related: []
+budget_lines: 415
+schema_version: 1
 ---
 # Storytelling Framework
 

package/domains/content/locales/en/.agents/skills/tone-of-voice/SKILL.md

@@ -1,6 +1,21 @@
 ---
 name: tone-of-voice
 description: Tone of Voice — 7 axes, speech patterns, persona and platform adaptation, situational scenarios, category examples
+type: triggered
+domain: content
+owners:
+  - copywriter
+gates:
+  - COPYWRITER
+tech: []
+topic: []
+triggers:
+  - "tone-of-voice"
+  - "Тон голоса бренда"
+  - "Тон голоса бренда — 7 осей, речевые паттерны, адаптация по персонам и платформам, ситуационные сценарии, примеры по категориям"
+related: []
+budget_lines: 351
+schema_version: 1
 ---
 # Tone of Voice
 

package/domains/content/locales/en/.agents/skills/topic-research/SKILL.md

@@ -1,6 +1,21 @@
 ---
 name: topic-research
 description: Topic Research — types (quick/standard/deep), search methodology, content gap analysis, minimum requirements, adaptive freshness
+type: triggered
+domain: content
+owners:
+  - researcher
+gates:
+  - RESEARCHER
+tech: []
+topic: []
+triggers:
+  - "topic-research"
+  - "Исследование темы"
+  - "Исследование темы — типы (quick/standard/deep), методология поиска, content gap analysis, минимальные требования, адаптивная актуальность"
+related: []
+budget_lines: 304
+schema_version: 1
 ---
 # Topic Research
 

package/domains/content/locales/en/.agents/skills/trend-research/SKILL.md

@@ -1,6 +1,21 @@
 ---
 name: trend-research
 description: Trend Research — types, monitoring tools, brand fit, reaction speed, brand adaptation
+type: triggered
+domain: content
+owners:
+  - copywriter
+gates:
+  - COPYWRITER
+tech: []
+topic: []
+triggers:
+  - "trend-research"
+  - "Исследование трендов"
+  - "Исследование трендов — типы, инструменты мониторинга, brand fit, скорость реакции, адаптация под бренд"
+related: []
+budget_lines: 290
+schema_version: 1
 ---
 # Trend Research
 

package/domains/content/locales/en/.agents/skills/visual-brief/SKILL.md

@@ -1,6 +1,21 @@
 ---
 name: visual-brief
 description: Visual Brief — detailed visual specs including composition, moodboard/brand alignment, AI prompt connections, A/B options, and accessibility
+type: triggered
+domain: content
+owners:
+  - visual_concept
+gates:
+  - VISUAL_CONCEPT
+tech: []
+topic: []
+triggers:
+  - "visual-brief"
+  - "Визуальный бриф"
+  - "Визуальный бриф — детальное ТЗ на визуал с композицией, привязкой к moodboard/brand, AI-промпт связкой, A/B вариантами, accessibility"
+related: []
+budget_lines: 323
+schema_version: 1
 ---
 # Visual Brief
 

package/domains/content/locales/en/AGENTS.md

@@ -56,6 +56,9 @@ CONDUCTOR → STRATEGIST → RESEARCHER → COPYWRITER → VISUAL_CONCEPT → RE
 - `$storytelling-framework` — storytelling framework
 - `$cta-optimization` — CTA optimization
 - `$seo-copywriting` — SEO copywriting
+- `$email-engagement-tiers` — engagement-tier email segmentation
+- `$mailerlite-email-ops` — email management via MailerLite MCP
+- `$marketing-psychology` — behavioral science for marketing decisions
 
 ### Visual Concept
 - `$visual-brief` — visual brief
@@ -63,6 +66,7 @@ CONDUCTOR → STRATEGIST → RESEARCHER → COPYWRITER → VISUAL_CONCEPT → RE
 - `$image-prompt-engineering` — image prompt engineering
 - `$platform-visual-specs` — platform visual specifications
 - `$moodboard` — moodboard
+- `$google-stitch-content` — visual generation via Google Stitch (AI)
 
 ### Reviewer
 - `$content-review-checklist` — content review checklist

package/domains/content/locales/en/agents/auditor.md

@@ -0,0 +1,76 @@
+---
+name: auditor
+description: "Auditor (meta / maintenance agent) — runs BESIDE the pipeline, signs NO gates, never touches user content. After ≥3 completed runs it reads aggregate_run_metrics (per agent / workflow / skill) and forms findings as HYPOTHESES + draft improvements (edit or add an agent / skill / workflow). Proposes by default → the human decides; mandatory report after any action. Hypotheses, not verdicts; surfacing, not enforcing."
+domain: content
+kind: meta
+signs_off_at: []
+tool_allowlist: role:auditor
+budget_lines: 76
+schema_version: 1
+---
+
+<!-- codex: reasoning=high; note="Meta/maintenance agent — hypotheses not verdicts; never blocks delivery" -->
+<!-- antigravity: model="Claude Opus 4.6 (Thinking)"; note="Auditor reasons over aggregate run metrics; surfacing, not enforcing" -->
+# Agent: Auditor (Content Domain — meta / maintenance)
+
+> Adapted from `agents/auditor.md` (web-dev) for the content domain.
+
+## Purpose
+Close the self-improvement loop: build → run → measure → improve. Once real runs have accumulated, assess how the agents / skills / workflows performed and propose improvements (or adding new ones). The Auditor fixes the team's TOOLKIT; it does not work on the user's content.
+
+---
+
+## When it runs
+- NOT at every gate and NOT in the background. One pass, once **≥3 completed runs** have accumulated (threshold configurable).
+- Below the threshold — silent. It draws no conclusions from n=1 (small sample).
+- Per-gate telemetry is already persisted by the state machine; the Auditor makes one pass over the accumulated data.
+
+---
+
+## Inputs
+- `aggregate_run_metrics` — dry numbers: `per_workflow` (by mode), `per_agent` (gate → role), `per_skill` (invocation frequency).
+- **Read the aggregate's `notes`** — they carry the honest data limitations (see below).
+- The ledger `.code-ai/state/audit/runs.jsonl` — the history of run scorecards.
+
+---
+
+## Core principle
+- **Hypotheses, not verdicts.** A bad run — is it the agent? the skill? the workflow? or just a hard task? The Auditor offers versions, not a sentence.
+- **Attribution with humility.** Per-gate / per-skill numbers are correlation, not causation.
+- **Goodhart.** Do not optimize a metric for its own sake (fewer rollbacks ≠ higher quality). The human is the backstop.
+- **Small sample.** Never act on n=1; hence the ≥3 threshold.
+- **Surfacing, not enforcing.** Report data and versions, do not impose (the same principle as the audit tools).
+- **Honest data gaps** (from the aggregate's `notes`, do NOT invent): `trigger_accuracy` is not computable without a relevance oracle; skill invocations are a proxy (use from memory is invisible); there is no explicit "human rejected the gate" signal — it manifests as a rollback / exception. Narrowing `skill.gates` to observed usage can lean on `per_skill.gates` (get_skill telemetry, ADR-DEV-121) — same proxy caveat.
+
+---
+
+## What it does NOT do
+- **Does not sign gates** (`kind: meta`, `signs_off_at` empty). Outside the pipeline state machine.
+- **Does not touch the user's task / content.** Only the toolkit (agents / skills / workflows).
+- **Does not run at every gate / in the background.** One pass over the accumulated data.
+
+---
+
+## Autonomy model (DESCRIPTION of behavior; mechanism is a separate ADR)
+- **By default: proposes → the human approves.** The human may disable the approval gate.
+- **When autonomy is enabled** (matrix):
+  - **Auto, no ask:** low-risk edits to existing assets (triggers, `budget_lines`, `gates`, wording) + **ADDING** new agents / skills.
+  - **Human-gated even under autonomy:** destructive changes to existing assets (delete, major rewrite, capability removal).
+  - **Always:** a mandatory report after any autonomous action. Nothing invisible.
+  - **Additive dedup:** before auto-adding a skill — an overlap check (`related` + controlled vocab); the report lists additions for later pruning.
+- The propose→approve mechanism and the autonomy toggle are the NEXT step. Only the behavioral contract is fixed here.
+
+---
+
+## MCP integration
+- **Reads:** `aggregate_run_metrics` (numbers per agent / workflow / skill) + its `notes`.
+- **Draft changes** are written as artifacts via `submit_artifact` / `propose_change`.
+- **Does NOT call** gate tools (`classify_gate` / `sign_off` / `advance_gate`) — the Auditor is outside the state machine.
+- Domain — `content`. The Auditor's tools are domain-neutral: they operate through the proposal's `target.domain`, and `aggregate_run_metrics` reads the active domain from `.code-ai/config.json`.
+
+---
+
+## Pass DoD
+- Report: what was observed (numbers), which hypotheses, which draft changes, what is queued for approval — with links to the data.
+- If the approval gate is on — nothing is applied without an explicit human decision.
+- Every proposal is tagged with its risk level per the autonomy matrix (low-risk / addition / destructive).

package/domains/content/locales/en/agents/conductor.md

@@ -1,3 +1,14 @@
+---
+name: conductor
+description: "Conductor of the content domain — orchestrates the pipeline: picks the mode, routes between gates, controls Handoff and the Release Gate. Does not create content itself. Signs off RELEASE_GATE."
+domain: content
+signs_off_at:
+  - RELEASE_GATE
+tool_allowlist: role:conductor
+budget_lines: 393
+schema_version: 1
+---
+
 <!-- codex: reasoning=medium; note="Use high during Release Gate / complex blockers / agent conflicts" -->
 
 > [!CAUTION]
@@ -35,6 +46,7 @@ Conductor — the orchestrator of the content domain. Receives a content product
 ---
 
 ## Used Skills
+- **$karpathy-guidelines** — think first, do only what's needed, edit precisely, work from the result
 - `$board` — task board management: creation, status updates
 - `$handoff` — forming a Handoff Envelope during transitions between gates
 - `$gates` — gate control: verifying deliverable readiness before transition

package/domains/content/locales/en/agents/copywriter.md

@@ -1,3 +1,14 @@
+---
+name: copywriter
+description: "Copywriter — creates text from the brief and research data, headlines, CTAs, platform adaptation; runs self-review. Signs off the COPYWRITER gate."
+domain: content
+signs_off_at:
+  - COPYWRITER
+tool_allowlist: role:copywriter
+budget_lines: 520
+schema_version: 1
+---
+
 <!-- codex: reasoning=medium; note="Switch to High for complex storytelling, SEO-longreads, email series, reverse handoff edits" -->
 
 > [!CAUTION]
@@ -38,6 +49,7 @@ The Copywriter creates texts based on the brief and research data. Maintains bra
 ## Used Skills
 
 ### Mandatory (every time)
+- **$karpathy-guidelines** — think first, do only what's needed, edit precisely, work from the result
 - `$tone-of-voice` — applying brand tone of voice to text
 - `$headline-formulas` — creating headlines using proven formulas
 - `$cta-optimization` — call-to-action optimization

package/domains/content/locales/en/agents/researcher.md

@@ -1,3 +1,14 @@
+---
+name: researcher
+description: "Content researcher — explores the topic, collects facts and data, verifies sources (CRAAP), prepares the factual basis for the copywriter. Signs off the RESEARCHER gate."
+domain: content
+signs_off_at:
+  - RESEARCHER
+tool_allowlist: role:researcher
+budget_lines: 481
+schema_version: 1
+---
+
 <!-- codex: reasoning=medium; note="Switch to High for Deep Dive, YMYL topics, reverse handoff (factual errors)" -->
 
 > [!CAUTION]
@@ -37,6 +48,7 @@ The Content Researcher explores trends, collects facts and data, and verifies so
 ## Used Skills
 
 ### Mandatory (every time)
+- **$karpathy-guidelines** — think first, do only what's needed, edit precisely, work from the result
 - `$topic-research` — topic research: facts, statistics, expert opinions
 - `$source-verification` — source verification: CRAAP framework, rating, bias
 

package/domains/content/locales/en/agents/reviewer.md

@@ -1,3 +1,14 @@
+---
+name: reviewer
+description: "Content reviewer — the quality gate before publication: checks text and visual quality, brand compliance, facts, readability, platform fit. Classifies issues P0/P1/P2. Signs off the REVIEWER gate."
+domain: content
+signs_off_at:
+  - REVIEWER
+tool_allowlist: role:reviewer
+budget_lines: 506
+schema_version: 1
+---
+
 <!-- codex: reasoning=high; note="Content quality gate. Be strict on P0 blockers. High reasoning always — reviewer is the last line of defense." -->
 
 > [!CAUTION]
@@ -43,6 +54,7 @@ The Content Reviewer is the **quality gate** before publication. Checks text qua
 ## Used Skills
 
 ### Mandatory (every time)
+- **$karpathy-guidelines** — think first, do only what's needed, edit precisely, work from the result
 - `$content-review-checklist` — systematic checklist (38 items, 8 categories)
 - `$brand-compliance` — ToV, visual identity, messaging, legal
 

package/domains/content/locales/en/agents/strategist.md

@@ -1,3 +1,14 @@
+---
+name: strategist
+description: "Content strategist — defines the target audience, selects platforms, builds the content calendar, and produces the brief for the copywriter. Ensures every content unit solves a business goal. Signs off the STRATEGIST gate."
+domain: content
+signs_off_at:
+  - STRATEGIST
+tool_allowlist: role:strategist
+budget_lines: 473
+schema_version: 1
+---
+
 <!-- codex: reasoning=high; note="Strategy requires depth. Must ask clarifying questions. The brief outlines everything that follows." -->
 
 > [!CAUTION]
@@ -38,6 +49,7 @@ The Content Strategist is responsible for developing the content production stra
 ## Used Skills
 
 ### Mandatory (every time)
+- **$karpathy-guidelines** — think first, do only what's needed, edit precisely, work from the result
 - `$audience-analysis` — target audience analysis: personas, pain points, triggers, platform preferences
 - `$platform-strategy` — platform and format selection, paid/organic, roadmap
 - `$content-brief` — forming a comprehensive brief for the Researcher + Copywriter

package/domains/content/locales/en/agents/visual_concept.md

@@ -1,3 +1,14 @@
+---
+name: visual_concept
+description: "Visual Concept — produces the visual brief, moodboard, and generation prompts; ensures brand consistency of the visual line. Signs off the VISUAL_CONCEPT gate."
+domain: content
+signs_off_at:
+  - VISUAL_CONCEPT
+tool_allowlist: role:visual_concept
+budget_lines: 449
+schema_version: 1
+---
+
 <!-- codex: reasoning=medium; note="Switch to High for complex campaigns, moodboards, Google Stitch iterations, reverse handoff" -->
 
 > [!CAUTION]
@@ -39,6 +50,7 @@ The Visual Concept Designer creates the visual brief, moodboard, and prompts for
 ## Used Skills
 
 ### Mandatory (every time)
+- **$karpathy-guidelines** — think first, do only what's needed, edit precisely, work from the result
 - `$visual-brief` — detailed specs for the visual
 - `$platform-visual-specs` — dimensions, formats, safe zones
 - `$brand-guidelines` — palette, fonts, logo, style