iron-edge 1.2.0 → 2.0.0

package/squads/squad-claude/knowledge-base/chrome-brain-integration.md

@@ -1,49 +1,24 @@
-# Chrome Brain Integration — Squad Claude (Claude Code Mastery)
-> Squad Claude tem acesso ao Chrome Brain para documentacao de novos patterns de automacao, solucoes NSN, discovery de MCP tools, gestao de configuracao e otimizacao de prompts para tarefas de browser.
-## Tier: 1
-## Quando Ativar
-- Descoberta de novo pattern de automacao que precisa ser documentado e testado
-- Solucao NSN encontrada que precisa ser validada em browser real antes de registrar
-- Discovery de novos MCP tools ou capabilities via browser (docs, changelogs, APIs)
-- Configuracao de MCP servers ou Chrome DevTools que requer teste interativo
-- Otimizacao de prompts para tarefas de browser (testar variantes, medir resultados)
-- Debug de integracao Chrome Brain com outros squads (reproduzir problema no browser)
-## O Que Recebe do Chrome Brain
-- Resultados de testes de novos patterns de automacao (funciona/nao funciona)
-- Screenshots de documentacao oficial (Claude, MCP, Chrome DevTools Protocol)
-- Changelogs e release notes extraidos de sites de ferramentas
-- Configuracoes testadas e validadas (JSON, YAML, env vars)
-- Metricas de performance de diferentes abordagens de automacao
-- Error logs e stack traces capturados durante debug de integracoes
-## O Que Envia pro Chrome Brain
-- Novos patterns validados para registro no Learnings Log
-- Scripts de teste para validacao de configuracoes MCP
-- Documentacao de solucoes NSN para propagacao cross-squad
-- Prompts otimizados para tarefas de browser recorrentes
-- Regras de auto-activation atualizadas baseadas em novos patterns
+# Navegador, quando disponível
 
-## Tools Disponiveis
+> ⚠️ **Nada aqui vem no pacote. Verifique antes de prometer ao usuário.**
 
-1. **Chrome DevTools MCP** (29 tools) — acoes rapidas, screenshots, audit
-2. **dev-browser** — scraping complexo, batch operations
-3. **claude-in-chrome** — fallback visual
+Alguns fluxos ficam melhores com um navegador controlável — conferir uma página no ar,
+tirar screenshot, ler o console. Isso **depende do que está instalado na máquina de quem
+usa**, e o pacote não instala nada disso.
 
-## Session Management
+## O que pode existir
 
-- Max 15 screenshots por sessao
-- Salvar outputs em arquivo antes de acumular
-- Rotacionar sessao a cada ~12 screenshots
+| Ferramenta | Como conferir |
+|---|---|
+| **Chrome DevTools MCP** | é um servidor MCP; confira se aparece na lista de ferramentas da sessão |
+| Qualquer outro MCP de navegador | idem — o nome varia por instalação |
 
-## NSN Mode
+## A regra
 
-Ativo. Se uma acao de browser falhar, tentar alternativas automaticamente.
-Buscar solucao na internet. Configurar, testar, repetir. Max 5 ciclos.
+**Antes de dizer que vai abrir o navegador, confirme que a ferramenta está na sessão.**
+Se não estiver, diga em uma linha o que não dá para fazer e siga com o que dá.
 
-## Auto-Learning
-
-Ao descobrir novo pattern ou resolver barreira via NSN, atualizar:
-`~/.sinapse/sinapse/knowledge-base/chrome-brain.md` → secao Learnings Log
-
-## Referencia Completa
-
-`~/.sinapse/sinapse/knowledge-base/chrome-brain.md`
+Até 22/08/2026 este arquivo afirmava que o squad "tem acesso" a um conjunto fixo de
+ferramentas, incluindo uma que só existia na máquina de quem escreveu o pacote. Era a
+mesma classe de defeito pela qual a versão 1.2.0 foi depreciada: capacidade prometida
+que não existe do outro lado.

package/squads/squad-claude/knowledge-base/claude-code-configuration-reference.md

@@ -50,7 +50,7 @@ Priority (highest to lowest):
       "Write(~/.ssh/*)",
       "Write(.env*)"
     ],
-    "auto_approve": [
+    "deny": [
       "Read",
       "Glob",
       "Grep",
@@ -73,8 +73,8 @@ Priority (highest to lowest):
 
 #### Permission Precedence
 
-1. `deny` always wins over `allow` and `auto_approve`
-2. `auto_approve` wins over `allow` (no prompt shown)
+1. `deny` always wins over `allow` and `deny`
+2. `deny` wins over `allow` (no prompt shown)
 3. `allow` is the default for unmatched tools (prompt shown)
 4. Unmatched tools prompt the user for approval
 
@@ -110,12 +110,7 @@ Priority (highest to lowest):
 {
   "hooks": {
     "PreToolUse": [
-      {
-        "matcher": "regex_pattern",
-        "command": "shell_command",
-        "timeout": 5000,
-        "on_error": "warn"
-      }
+      { "matcher": "Bash", "hooks": [ { "type": "command", "command": "shell_command", "timeout": 30 } ] }
     ],
     "PostToolUse": [],
     "Notification": [],
@@ -221,7 +216,7 @@ Managed policies are distributed via:
 ```
 Managed deny + Project allow = DENY (managed always wins)
 Managed allow + Project deny = DENY (most restrictive wins for deny)
-Managed auto_approve + Project deny = DENY (deny always wins)
+Managed deny + Project deny = DENY (deny always wins)
 ```
 
 ---
@@ -306,7 +301,7 @@ Start with restrictive permissions and open up as needed:
       "Write(*.key)",
       "Read(.env*)"
     ],
-    "auto_approve": [
+    "deny": [
       "Read(src/**)",
       "Read(tests/**)",
       "Glob",

package/squads/squad-claude/knowledge-base/claude-code-hooks-reference.md

@@ -16,24 +16,43 @@ Hooks can be defined at multiple levels:
 ```json
 {
   "hooks": {
-    "EventName": [
+    "PreToolUse": [
       {
-        "matcher": "optional_pattern",
-        "command": "shell_command_to_execute",
-        "timeout": 5000,
-        "on_error": "warn"
+        "matcher": "Bash",
+        "hooks": [
+          { "type": "command", "command": "~/.claude/hooks/meu-hook.sh", "timeout": 30 }
+        ]
       }
     ]
   }
 }
 ```
 
+⚠️ **A forma é ANINHADA, e isso não é detalhe.** A entrada do array tem `matcher` e `hooks`;
+o comando fica DENTRO do array `hooks`, num objeto com `type` e `command`. `timeout` vive
+nesse objeto interno. **Não existe campo `on_error`** — conferido em 22/08/2026 contra
+https://code.claude.com/docs/en/hooks e contra um `settings.json` real em funcionamento.
+
+Até essa data este arquivo ensinava a forma PLANA (`{matcher, command, timeout, on_error}`),
+
+inexistentes. **Se você copiou daqui antes, seu hook não estava rodando.**
+
+### Os eventos que existem
+
+SessionStart · Setup · UserPromptSubmit · UserPromptExpansion · PreToolUse ·
+PermissionRequest · PermissionDenied · PostToolUse · PostToolUseFailure · PostToolBatch ·
+Notification · MessageDisplay · SubagentStart · SubagentStop · TaskCreated · TaskCompleted ·
+Stop · StopFailure · TeammateIdle · InstructionsLoaded · ConfigChange · CwdChanged ·
+DirectoryAdded · FileChanged · WorktreeCreate · WorktreeRemove · PreCompact · PostCompact ·
+Elicitation · ElicitationResult · SessionEnd
+
+
+
 | Field | Type | Required | Description |
 |-------|------|----------|-------------|
 | `matcher` | string (regex) | No | Pattern to match against the tool name or event context. If omitted, hook fires for ALL events of this type. |
 | `command` | string | Yes | Shell command to execute. Receives event payload via stdin as JSON. |
 | `timeout` | number (ms) | No | Maximum execution time. Default varies by event type. |
-| `on_error` | string | No | Behavior on command failure: `"warn"` (log and continue), `"error"` (block and report), `"ignore"` (silent). Default: `"warn"`. |
 
 ---
 
@@ -78,7 +97,6 @@ Hooks can be defined at multiple levels:
         "matcher": "Bash",
         "command": "python .claude/hooks/block-dangerous-git.py",
         "timeout": 3000,
-        "on_error": "error"
       }
     ]
   }
@@ -128,7 +146,6 @@ Hooks can be defined at multiple levels:
         "matcher": "Write|Edit",
         "command": "node .claude/hooks/auto-lint.js",
         "timeout": 10000,
-        "on_error": "warn"
       }
     ]
   }
@@ -167,7 +184,6 @@ Hooks can be defined at multiple levels:
       {
         "command": "python .claude/hooks/desktop-notify.py",
         "timeout": 5000,
-        "on_error": "ignore"
       }
     ]
   }
@@ -209,7 +225,6 @@ Hooks can be defined at multiple levels:
       {
         "command": "node .claude/hooks/track-tokens.js",
         "timeout": 5000,
-        "on_error": "ignore"
       }
     ]
   }
@@ -240,7 +255,7 @@ Hooks can be defined at multiple levels:
 - Clean up subagent worktrees
 - Validate subagent output before accepting
 
-### PreToolUseOutput
+
 
 **Trigger:** Before tool output is presented to the model.
 **Purpose:** Filter, transform, or redact tool output before Claude processes it.
@@ -252,7 +267,7 @@ Hooks can be defined at multiple levels:
 - Transform output format for better model comprehension
 - Add annotations or metadata to output
 
-### PostToolUseOutput
+
 
 **Trigger:** After tool output has been processed by the model.
 **Purpose:** Log final tool results, update external tracking, chain reactions.
@@ -279,7 +294,6 @@ Chain PostToolUse with a linter to automatically fix code after generation:
         "matcher": "Write|Edit",
         "command": "bash -c 'INPUT=$(cat); FILE=$(echo $INPUT | jq -r .tool_input.file_path); npx eslint --fix \"$FILE\" 2>/dev/null || true'",
         "timeout": 15000,
-        "on_error": "warn"
       }
     ]
   }
@@ -298,7 +312,6 @@ Block execution of dangerous commands using PreToolUse:
         "matcher": "Bash",
         "command": "bash -c 'INPUT=$(cat); CMD=$(echo $INPUT | jq -r .tool_input.command); echo \"$CMD\" | grep -qE \"rm -rf /|:(){ :|:& };:|dd if=/dev|mkfs|> /dev/sd\" && exit 1 || exit 0'",
         "timeout": 2000,
-        "on_error": "error"
       }
     ]
   }
@@ -316,7 +329,6 @@ Route notifications to different channels based on severity:
       {
         "command": "python .claude/hooks/route-notification.py",
         "timeout": 10000,
-        "on_error": "ignore"
       }
     ]
   }
@@ -337,7 +349,6 @@ Run relevant tests after code changes:
         "matcher": "Write|Edit",
         "command": "bash -c 'INPUT=$(cat); FILE=$(echo $INPUT | jq -r .tool_input.file_path); if [[ \"$FILE\" == *.test.* ]]; then npx jest \"$FILE\" --passWithNoTests 2>&1 | tail -5; elif [[ \"$FILE\" == *.ts || \"$FILE\" == *.tsx ]]; then TEST=\"${FILE%.ts*}.test${FILE##*.ts}\"; [ -f \"$TEST\" ] && npx jest \"$TEST\" --passWithNoTests 2>&1 | tail -5; fi'",
         "timeout": 30000,
-        "on_error": "warn"
       }
     ]
   }
@@ -356,7 +367,6 @@ Run quality checks before any git commit:
         "matcher": "Bash",
         "command": "bash -c 'INPUT=$(cat); CMD=$(echo $INPUT | jq -r .tool_input.command); if echo \"$CMD\" | grep -q \"git commit\"; then npm run lint --silent && npm run typecheck --silent || (echo \"Quality checks failed\" && exit 1); fi'",
         "timeout": 60000,
-        "on_error": "error"
       }
     ]
   }
@@ -374,7 +384,6 @@ Track cumulative token usage and warn when approaching limits:
       {
         "command": "node .claude/hooks/token-budget.js",
         "timeout": 3000,
-        "on_error": "ignore"
       }
     ]
   }
@@ -392,7 +401,7 @@ Track cumulative token usage and warn when approaching limits:
 - Use `timeout` to prevent hung hooks from blocking the session
 
 ### Reliability
-- Always set `on_error` explicitly — don't rely on defaults
+- O `timeout` vive DENTRO do objeto do array `hooks`, junto de `type` e `command` — nao no nivel do `matcher`.
 - Use `"error"` only for security-critical PreToolUse gates
 - Use `"warn"` for automation that should not block work
 - Use `"ignore"` for optional enhancements (notifications, logging)

package/squads/squad-claude/knowledge-base/claude-code-internals-deep.md

@@ -1,5 +1,19 @@
 # Claude Code Internals — Deep Reference
 
+> ⚠️ **NADA NESTE ARQUIVO ESTÁ NA DOCUMENTAÇÃO OFICIAL, E ELE NÃO SERVE COMO REFERÊNCIA.**
+>
+> Verificado em 22/08/2026 contra https://code.claude.com/docs: os sistemas descritos aqui
+> — incluindo os nomes de projeto internos, as contagens de feature flag, a sequência de
+> boot e os vetores de cache — **não aparecem em nenhuma página pública**. O próprio arquivo
+> declara vir de engenharia reversa de um pacote publicado.
+>
+> Consequências, ditas com todas as letras: pode estar errado, pode ter sido inventado,
+> pode descrever coisa que nunca existiu, e certamente descreve coisa que muda sem aviso.
+> **Não decida arquitetura, não escreva agente e não ensine ninguém com base nisto.**
+> Para o que é real e suportado, use a documentação oficial.
+
+
+
 > Source: Analysis from source code leak (npm v2.1.88, March 31 2026). 512K+ lines TypeScript, 1,902 files, 35 subsystems, 184 tools, 44 feature flags.
 
 ---
@@ -565,9 +579,9 @@ $0, $1 for indexed arguments
 
 ---
 
-## SINAPSE Architectural Lessons
+## Iron Edge Architectural Lessons
 
-From the source code analysis, 7 lessons for SINAPSE:
+From the source code analysis, 7 lessons for Iron Edge:
 
 1. **Minimal Core Loop, Maximum Infrastructure** — keep orchestration loop simple (~20 lines), invest in tools and context management
 2. **Safety Through Proximity** — embed constraints in tool descriptions, not separate policy files

package/squads/squad-claude/knowledge-base/claude-code-tools-reference.md

@@ -247,7 +247,7 @@ ENABLE_TOOL_SEARCH=false
 
 "Safety rules appear in the immediate context of the action they govern" — embedded directly in tool descriptions rather than separate policy files. This makes them harder for LLMs to overlook.
 
-**Applied in SINAPSE:** Hook governance rules embedded in tool-specific hook scripts, not just in documentation.
+**Applied in Iron Edge:** Hook governance rules embedded in tool-specific hook scripts, not just in documentation.
 
 ### Tool Authority Matrix
 
@@ -292,7 +292,7 @@ MCP tools are deferred by default — only tool names consume context until used
 
 ---
 
-## SINAPSE Tool Selection Guidance
+## Iron Edge Tool Selection Guidance
 
 ### When to use each approach
 

package/squads/squad-claude/knowledge-base/context-window-optimization.md

@@ -250,7 +250,7 @@ Parent system prompt (cached):
 
 ### Agent Memory Context Budget
 
-For SINAPSE agents, recommended allocation:
+For Iron Edge agents, recommended allocation:
 
 ```
 200K token session:
@@ -306,7 +306,7 @@ Research finding: LLMs show significantly degraded performance when critical inf
 
 ---
 
-## SINAPSE Context Architecture
+## Iron Edge Context Architecture
 
 ### System Prompt Allocation
 
@@ -320,7 +320,7 @@ Total baseline:                   ~10K tokens
 
 ### Agent Memory Allocation
 
-Each SINAPSE agent file: target < 300 lines / ~3K tokens.
+Each Iron Edge agent file: target < 300 lines / ~3K tokens.
 
 **Why:** Agents are loaded into context when activated. Large agent definitions increase baseline cost for every session turn.
 

package/squads/squad-claude/knowledge-base/knowledge-architecture-reference.md

@@ -369,11 +369,11 @@ Official [obsidian-skills](https://github.com/kepano/obsidian-skills) (by Obsidi
 
 ---
 
-## SINAPSE Knowledge Architecture
+## Iron Edge Knowledge Architecture
 
 ### Current State
 
-SINAPSE uses a document-first approach:
+Iron Edge uses a document-first approach:
 - Stories in `docs/stories/`
 - Architecture in `docs/architecture/`
 - KB files in `squads/{squad}/knowledge-base/`
@@ -384,7 +384,7 @@ SINAPSE uses a document-first approach:
 2. **Frontmatter standards** — Add `created_date`, `updated_date`, `review_by`, `status`, `links_to` to KB files
 3. **Quality scoring** — Periodic audit of KB files using quality formula
 4. **Decay detection** — Flag KB files not updated in 90+ days
-5. **Graph visualization** — Use `sinapse graph` to visualize knowledge connections
+5. **Graph visualization** — Use `iron-edge graph` to visualize knowledge connections
 
 ### KB File Template
 

package/squads/squad-claude/knowledge-base/mcp-integration-patterns.md

@@ -260,7 +260,7 @@ curl -X POST http://localhost:8080/servers \
 
 | Issue | Cause | Fix |
 |-------|-------|-----|
-| "Server not found" | MCP config not in settings.json | Add server config to correct settings file |
+| "Server not found" | MCP nao fica em settings.json | Add server config to correct settings file |
 | "Connection refused" | Server not running / wrong port | Start server, check URL/port |
 | "Tool not found" | Server connected but tool not registered | Check server implements the tool, restart server |
 | "Authentication failed" | Invalid or expired token | Rotate token, check env var name |

package/squads/squad-claude/knowledge-base/memory-systems-reference.md

@@ -101,9 +101,9 @@ from mem0 import Memory
 
 m = Memory()
 # Store memory
-m.add("User prefers dark mode", user_id="caio")
+m.add("User prefers dark mode", user_id="usuario")
 # Search memory
-results = m.search("user preferences", user_id="caio")
+results = m.search("user preferences", user_id="usuario")
 ```
 
 ### A-Mem (NeurIPS 2025)
@@ -339,7 +339,7 @@ class AgentMemory:
 ```python
 # Track how facts change over time
 fact = KnowledgeFact(
-    content="User works at Astro Brand Studio",
+    content="User works at <sua empresa>",
     valid_from="2025-01",
     valid_to=None,          # Still valid
     ingested_at="2025-01-15",
@@ -389,11 +389,11 @@ Without governance, knowledge systems accumulate ROT (Redundant, Outdated, Trivi
 
 ---
 
-## SINAPSE Memory Integration
+## Iron Edge Memory Integration
 
 ### Current Architecture
 
-Claude Code's built-in memory (`~/.claude/agent-memory/sinapse-claude/`) follows the 3-layer pattern:
+Claude Code's built-in memory (`~/.claude/projects/<slug-do-projeto>/memory/`) follows the 3-layer pattern:
 
 ```
 MEMORY.md (index, always loaded)
@@ -405,7 +405,7 @@ MEMORY.md (index, always loaded)
 
 ### Recommended Enhancements
 
-1. **AutoDream-equivalent:** Periodic consolidation of memory files (SINAPSE could run this as a daily hook)
+1. **AutoDream-equivalent:** Periodic consolidation of memory files (Iron Edge could run this as a daily hook)
 2. **Quality scoring:** Add `quality_score` frontmatter to memory files
 3. **Decay tracking:** Add `review_by` dates to time-sensitive memories
 4. **Team memory:** Shared memory across agents using `teamMemPaths.ts` pattern

package/squads/squad-claude/knowledge-base/obsidian-claude-integration.md

@@ -196,7 +196,7 @@ project: "Sprint 23"
 story_id: "2.3"
 status: "InProgress"
 priority: "High"
-assignee: "Caio"
+assignee: "voce"
 due_date: 2026-04-15
 ```
 
@@ -386,11 +386,11 @@ git push origin main --quiet
 
 ---
 
-## Caio's Vault Architecture
+## Arquitetura de cofre — exemplo de referencia
 
 ### Current Configuration
 
-**Path:** `C:\Users\Caio Imori\OneDrive\Caio Imori\3. B U S I S S N E S\ClaudeCode\#AI-Projects\Second-Brain`
+**Path:** `C:\Users\<voce>\OneDrive\<pasta de negocios>\ClaudeCode\AI-Projects\Second-Brain`
 
 **Routing:** `~/.claude/vault-routing.json` maps project domains to vault notes.
 
@@ -400,8 +400,8 @@ git push origin main --quiet
 
 | Domain | Vault Note | What It Covers |
 |--------|-----------|----------------|
-| SINAPSE | sinapse-ai-framework.md, sinapse.md | Framework, course, platform |
-| Astro Brand Studio | astro-brand-studio.md | Agency, services, clients |
+| Iron Edge | iron-edge-framework.md, iron-edge.md | Framework, course, platform |
+| Minha Agencia | minha-agencia.md | Agency, services, clients |
 | @caioimori | marca-pessoal-caioimori.md | Personal brand, content |
 | Colegio Modulo | colegio-modulo.md | Marketing, branding |
 

package/squads/squad-claude/knowledge-base/retrieval-augmented-generation.md

@@ -223,7 +223,7 @@ For each sub-question:
 
 **Survey:** arXiv 2501.09136 — "Agentic RAG" as formal research area.
 
-### SINAPSE Research Pipeline (analogous)
+### Iron Edge Research Pipeline (analogous)
 
 ```
 @analyst receives research request
@@ -275,7 +275,7 @@ results = evaluate(
 
 ---
 
-## RAG for SINAPSE Knowledge Base
+## RAG for Iron Edge Knowledge Base
 
 ### Current Knowledge Flow
 
@@ -288,7 +288,7 @@ User query → Claude Code
 ### Recommended Enhancement
 
 ```
-User query → SINAPSE agent
+User query → Iron Edge agent
   → Query formulation
   → Hybrid search over KB + stories + architecture docs
   → Top-N relevant chunks

package/squads/squad-claude/knowledge-base/skill-creation-patterns.md

@@ -242,22 +242,22 @@ The dominant community reference with 134K stars — not just skills, but a comp
 
 ---
 
-## Skills for SINAPSE Squad-Claude
+## Skills for Iron Edge Squad-Claude
 
 ### Existing Skills to Build
 
 | Skill | Priority | Description |
 |-------|----------|-------------|
 | `claude-code-audit` | High | Audit CC configuration, hooks, settings |
-| `sinapse-setup` | High | Initialize SINAPSE in a new project |
+| `iron-setup` | High | Initialize Iron Edge in a new project |
 | `agent-persona-creation` | High | Create new agent .md files |
 | `hooks-architecture` | Medium | Design and implement hook systems |
 | `mcp-server-setup` | Medium | Configure and test MCP servers |
 | `context-optimization` | Medium | Analyze and reduce context window usage |
-| `skill-creation` | High | Meta-skill — create new SINAPSE skills |
+| `skill-creation` | High | Meta-skill — create new Iron Edge skills |
 | `squad-publishing` | Low | Package and publish squads |
 
-### Skill Template for SINAPSE
+### Skill Template for Iron Edge
 
 ```yaml
 ---
@@ -318,16 +318,16 @@ Do NOT use when:
 
 ```json
 {
-  "name": "sinapse-claude-skills",
+  "name": "iron-edge-claude-skills",
   "plugins": [
     {
       "name": "development-workflow",
-      "description": "Skills for Claude Code configuration and SINAPSE workflows",
+      "description": "Skills for Claude Code configuration and Iron Edge workflows",
       "source": "./",
       "strict": false,
       "skills": [
         "./skills/claude-code-audit",
-        "./skills/sinapse-setup",
+        "./skills/iron-setup",
         "./skills/agent-persona-creation"
       ]
     }
@@ -338,15 +338,15 @@ Do NOT use when:
 ### Installation
 
 ```bash
-/plugin marketplace add sinapse-ai/squad-claude-skills
-/plugin install development-workflow@sinapse-claude-skills
+/plugin marketplace add iron-edge/squad-claude-skills
+/plugin install development-workflow@iron-edge-claude-skills
 ```
 
 ### Namespacing
 
 Skills from plugins receive prefix: `plugin-name:skill-name`
 
-Example: `sinapse-claude:claude-code-audit`
+Example: `iron-claude:claude-code-audit`
 
 ---
 

package/squads/squad-claude/knowledge-base/swarm-orchestration-patterns.md

@@ -165,9 +165,9 @@ Rules:
   |-- [Maintenance Agent] → Detects decay, archives, cleans up
 ```
 
-**For SINAPSE squads, this maps to:**
+**For Iron Edge squads, this maps to:**
 ```
-[sinapse-orqx / squad *-orqx]
+[iron-ceo / squad *-orqx]
   |-- @developer → implementation tasks
   |-- @quality-gate → validation tasks
   |-- @architect → design tasks
@@ -241,7 +241,7 @@ Generator Agent → produces initial output
   → If BLOCKED: Escalate to human
 ```
 
-Used in SINAPSE for story validation (sprint-lead creates → product-lead validates → developer implements → quality-gate reviews).
+Used in Iron Edge for story validation (sprint-lead creates → product-lead validates → developer implements → quality-gate reviews).
 
 ---
 
@@ -323,7 +323,7 @@ Workers request authorization for dangerous operations via a queue:
 
 ## BMAD Method Patterns
 
-BMAD (Breakthrough Method for Agile AI-Driven Development) v6 provides patterns SINAPSE can adopt.
+BMAD (Breakthrough Method for Agile AI-Driven Development) v6 provides patterns Iron Edge can adopt.
 
 ### Document Sharding
 
@@ -332,7 +332,7 @@ Large documents split into focused pieces:
 - Sharded: ~300 tokens per shard
 - **74-90% token consumption reduction**
 
-**Applied to SINAPSE PRDs:** Instead of one large PRD.md, shard into:
+**Applied to Iron Edge PRDs:** Instead of one large PRD.md, shard into:
 - `prd-overview.md` (~300 tokens)
 - `prd-requirements-fr.md` (~300 tokens)
 - `prd-requirements-nfr.md` (~300 tokens)
@@ -345,13 +345,13 @@ Agents load only the shards they need.
 
 Multiple agent personas collaborating within a single session — relevant context shared, irrelevant context excluded per persona.
 
-**SINAPSE equivalent:** Each `@agent` activation follows the handoff protocol, compacting previous agent to ~379 tokens.
+**Iron Edge equivalent:** Each `@agent` activation follows the handoff protocol, compacting previous agent to ~379 tokens.
 
 ---
 
 ## Guardrails and Governance
 
-### Authority Matrix (SINAPSE)
+### Authority Matrix (Iron Edge)
 
 | Agent | Can Write | Read-Only | Blocked |
 |-------|-----------|-----------|---------|
@@ -359,13 +359,13 @@ Multiple agent personas collaborating within a single session — relevant conte
 | @quality-gate | `tests/`, review files | Source code | Write to src/ |
 | @architect | `docs/architecture/` | System-wide | Application code |
 | @devops | Remote operations | Everything | — |
-| @sinapse-orqx | Everything | Everything | — |
+| @iron-ceo | Everything | Everything | — |
 
 ### Escalation Protocol
 
 ```
 Agent cannot complete task
-  → Escalate to @sinapse-orqx
+  → Escalate to @iron-ceo
   
 Quality gate fails
   → Return to @developer with specific feedback
@@ -374,7 +374,7 @@ Constitutional violation detected
   → BLOCK, require fix before proceeding
   
 Agent boundary conflict
-  → @sinapse-orqx mediates
+  → @iron-ceo mediates
 ```
 
 ### Infinite Loop Prevention