ostacky 0.6.3 → 0.7.0

package/assets/mcp/ostacky-controller/package.json

@@ -1,6 +1,6 @@
 {
     "name": "ostacky-controller",
-    "version": "0.6.3",
+    "version": "0.7.0",
     "private": true,
     "type": "module",
     "dependencies": {

package/assets/skills/brainstorming/SKILL.md

@@ -9,7 +9,7 @@ A thinking partner that adapts to what the user needs: structured design when th
 
 **Follow Core Instructions** — `ostacky.md` Core Instructions section for CodeGraph and Engram usage patterns.
 
-**IMPORTANT:** Engram is an **MCP server**, not a skill. Tools `mem_save`, `mem_search`, `mem_context` are MCP tools. Do NOT use `skill("engram")` — it doesn't exist.
+**IMPORTANT:** Engram is an **MCP server**, not a skill. Tools `engram_mem_save`, `engram_mem_search`, `engram_mem_context` are MCP tools. Do NOT use `skill("engram")` — it doesn't exist.
 
 ---
 
@@ -35,29 +35,29 @@ Do NOT invoke any implementation skill, write any code, or scaffold any project
 
 ### Process
 
-1. **Check Engram** — `mem_search` with keywords from the user's idea. Surface any prior design decisions or similar proposals.
-2. **Explore via CodeGraph** — `codegraph_explore` on the affected area. Only `Read` files CodeGraph didn't cover.
+1. **Check Engram** — `engram_mem_search` with keywords from the user's idea. Surface any prior design decisions or similar proposals.
+2. **Explore via CodeGraph** — `codegraph_codegraph_explore` on the affected area. Only `Read` files CodeGraph didn't cover.
 3. **Ask clarifying questions** — one at a time, understand purpose/constraints/success criteria
 4. **Propose 2-3 approaches** — with trade-offs and your recommendation
 5. **Present design** — in sections scaled to complexity, get user approval after each section
 6. **Write design doc** — save to `docs/superpowers/specs/YYYY-MM-DD-<topic>-design.md` and commit
 7. **Spec self-review** — check for placeholders, contradictions, ambiguity, scope
 8. **User reviews spec** — ask user to review before proceeding
-9. **Save to Engram** — `mem_save` with the design decision and tradeoffs
+9. **Save to Engram** — `engram_mem_save` with the design decision and tradeoffs
 10. **Transition** — based on routing decision (see Transition Rules below)
 
 ### Transition Rules
 
 The next step depends on how the change was routed by Ostacky:
 
-| Routing | Next Skill | When |
-|---------|------------|------|
-| **DIRECT** (Level 0/0+1) | `writing-plans` | Small changes, no OpenSpec |
+| Routing | Next Step | When |
+|---------|-----------|------|
+| **DIRECT** (Level 0/0+1) | **Implementación directa** (con gate de confirmación del usuario) | Small changes, no OpenSpec |
 | **SPEC** (Level 1+) | `openspec-propose` | Complex changes requiring OpenSpec artifacts |
 
 If you're unsure about routing, ask Ostacky or check the controller state.
 
-**Terminal state:** Invoke the appropriate next skill. Do NOT invoke implementation skills directly.
+**Terminal state:** For DIRECT, present the design and get user approval before proceeding to implementation. For SPEC, invoke `openspec-propose`. Do NOT invoke implementation skills directly.
 
 ### Design Principles
 

package/assets/skills/execution-mode-evaluation/SKILL.md

@@ -8,7 +8,7 @@ metadata:
   version: "2.0"
 ---
 
-**IMPORTANT:** Engram is an **MCP server**, not a skill. Tools `mem_save`, `mem_search`, `mem_context` are MCP tools. Do NOT use `skill("engram")` — it doesn't exist.
+**IMPORTANT:** Engram is an **MCP server**, not a skill. Tools `engram_mem_save`, `engram_mem_search`, `engram_mem_context` are MCP tools. Do NOT use `skill("engram")` — it doesn't exist.
 
 # Skill: execution-mode-evaluation
 
@@ -19,19 +19,19 @@ Determinar el modo de ejecución óptimo entre **inline** y **subagent-driven**
 | Dato | Fuente | Obligatorio |
 |------|--------|-------------|
 | Tasks del change | `tasks.md` del cambio activo | ✅ |
-| Archivos que modifica cada task | `codegraph_explore` o lectura directa | ✅ |
-| Blast radius por símbolo | `codegraph_impact` | Para alta precisión |
+| Archivos que modifica cada task | `codegraph_codegraph_explore` o lectura directa | ✅ |
+| Blast radius por símbolo | `codegraph_codegraph_impact` | Para alta precisión |
 | Contratos entre tasks | `design.md` | Para evaluar dependencias |
 
 ## Procedimiento
 
 ### Paso 0: Verificar datos existentes en contexto
 
-Si ya tenés output de `codegraph_explore` para el área del cambio **Y** ese output distingue archivos por task → **saltá al Paso 0.5**. Si no, ejecutá el Paso 1.
+Si ya tenés output de `codegraph_codegraph_explore` para el área del cambio **Y** ese output distingue archivos por task → **saltá al Paso 0.5**. Si no, ejecutá el Paso 1.
 
 ### Paso 0.1: Consultar Engram por decisiones previas
 
-`mem_search` con keywords del cambio (nombre del módulo, área afectada). Si existe una decisión de modo de ejecución anterior para un cambio similar, considerarla como referencia — no como vinculante. Las condiciones pueden haber cambiado.
+`engram_mem_search` con keywords del cambio (nombre del módulo, área afectada). Si existe una decisión de modo de ejecución anterior para un cambio similar, considerarla como referencia — no como vinculante. Las condiciones pueden haber cambiado.
 
 ### Paso 0.5: Early exit para cambios pequeños
 
@@ -51,7 +51,7 @@ Si el change tiene **≤2 tasks** Y **no comparten archivos entre sí** → devo
 ### Paso 1: Obtener datos de CodeGraph
 
 ```
-codegraph_explore con query: "<área del cambio>"
+codegraph_codegraph_explore con query: "<área del cambio>"
 ```
 
 Si el output es muy general → `codegraph_impact` sobre símbolos centrales para blast radius preciso.
@@ -110,7 +110,7 @@ Para cada fase de `tasks.md`, evaluar intra-fase:
 {
   "recommendation": "INLINE" | "SUBAGENT_DRIVEN",
   "reasons": ["razón principal", "razón secundaria"],
-  "codegraphUsed": ["codegraph_explore"],
+  "codegraphUsed": ["codegraph_codegraph_explore"],
   "taskCount": <N>,
   "sharedFiles": { "src/archivo.ts": ["task1", "task2"] },
   "fileClusters": [["task1", "task2"], ["task3"]],
@@ -178,7 +178,7 @@ Para cada fase de `tasks.md`, evaluar intra-fase:
 
 ## Checklist
 
-- [ ] Ejecuté `codegraph_explore` (o verifiqué datos existentes)?
+- [ ] Ejecuté `codegraph_codegraph_explore` (o verifiqué datos existentes)?
 - [ ] Construí mapa de dependencias con `fileClusters`?
 - [ ] Identifiqué clusters (componentes conectados)?
 - [ ] Verifiqué deps ENTRE clusters (no solo intra)?

package/assets/skills/graceful-degradation/SKILL.md

@@ -9,7 +9,7 @@ When critical tools are unavailable, this skill provides a systematic approach t
 
 **Follow Core Instructions** — `ostacky.md` Core Instructions section for CodeGraph and Engram usage patterns.
 
-**IMPORTANT:** Engram is an **MCP server**, not a skill. Tools `mem_save`, `mem_search`, `mem_context` are MCP tools. Do NOT use `skill("engram")` — it doesn't exist.
+**IMPORTANT:** Engram is an **MCP server**, not a skill. Tools `engram_mem_save`, `engram_mem_search`, `engram_mem_context` are MCP tools. Do NOT use `skill("engram")` — it doesn't exist.
 
 ---
 
@@ -45,7 +45,7 @@ Switch to basic mode with manual workflows.
 
 **Workflow:**
 ```
-1. mem_search for related analysis (if Engram available)
+1. engram_mem_search for related analysis (if Engram available)
 2. Use Glob to find relevant files: **/*.ts, src/**/*.ts
 3. Read files manually to understand structure
 4. Proceed with caution — no blast radius analysis
@@ -209,7 +209,7 @@ Voy a usar Engram para buscar análisis previos y luego leer archivos manualment
 
 User: Sí
 
-Agent: [Uses mem_search to find related analysis]
+Agent: [Uses engram_mem_search to find related analysis]
        [Uses Glob to find relevant files]
        [Reads files manually]
        [Proceeds with task]

package/assets/skills/openspec-apply-change/SKILL.md

@@ -9,7 +9,7 @@ metadata:
   generatedBy: "1.3.1"
 ---
 
-**IMPORTANT:** Engram is an **MCP server**, not a skill. Tools `mem_save`, `mem_search`, `mem_context` are MCP tools. Do NOT use `skill("engram")` — it doesn't exist.
+**IMPORTANT:** Engram is an **MCP server**, not a skill. Tools `engram_mem_save`, `engram_mem_search`, `engram_mem_context` are MCP tools. Do NOT use `skill("engram")` — it doesn't exist.
 
 Implement tasks from an OpenSpec change.
 
@@ -19,7 +19,7 @@ Implement tasks from an OpenSpec change.
 
 1. **Check Engram for past implementation patterns**
 
-   Before implementing, `mem_search` with keywords from the change name and affected modules. Look for:
+   Before implementing, `engram_mem_search` with keywords from the change name and affected modules. Look for:
    - Similar changes implemented before (approaches that worked or failed)
    - Known gotchas or edge cases in related areas
    - Model selection patterns for similar task complexity
@@ -62,7 +62,7 @@ Implement tasks from an OpenSpec change.
 
 5. **Read context files**
 
-   Use **CodeGraph** (`codegraph_explore`) to understand the code structure before reading files directly. Then read every file path listed under `contextFiles` from the apply instructions output.
+   Use **CodeGraph** (`codegraph_codegraph_explore`) to understand the code structure before reading files directly. Then read every file path listed under `contextFiles` from the apply instructions output.
    The files depend on the schema being used:
    - **spec-driven**: proposal, specs, design, tasks
    - Other schemas: follow the contextFiles from CLI output

package/assets/skills/openspec-propose/SKILL.md

@@ -9,7 +9,7 @@ metadata:
   generatedBy: "1.3.1"
 ---
 
-**IMPORTANT:** Engram is an **MCP server**, not a skill. Tools `mem_save`, `mem_search`, `mem_context` are MCP tools. Do NOT use `skill("engram")` — it doesn't exist.
+**IMPORTANT:** Engram is an **MCP server**, not a skill. Tools `engram_mem_save`, `engram_mem_search`, `engram_mem_context` are MCP tools. Do NOT use `skill("engram")` — it doesn't exist.
 
 Propose a new change - create the change and generate all artifacts in one step.
 
@@ -26,7 +26,7 @@ When ready to implement, run /opsx-apply
 
 **Steps**
 
-1. **Check Engram for similar changes** — `mem_search` with keywords from the user's description. If a similar change was proposed or implemented before, surface it to the user. They may want to extend the existing work instead of starting fresh.
+1. **Check Engram for similar changes** — `engram_mem_search` with keywords from the user's description. If a similar change was proposed or implemented before, surface it to the user. They may want to extend the existing work instead of starting fresh.
 
 2. **If no clear input provided, ask what they want to build**
 

package/assets/skills/review/SKILL.md

@@ -52,8 +52,8 @@ After obtaining the diff, if it is empty, inform the user that there are no chan
 
 Before analyzing the diff:
 
-1. **Follow Core Instructions** — `ostacky.md` Core Instructions section for CodeGraph usage patterns. Use `codegraph_explore` to understand the code structure before reading files directly.
-2. **Use CodeGraph for code understanding** — `codegraph_explore` on the changed symbols gives you call paths, blast radius, and related code in ONE call. Only `Read` files for details CodeGraph didn't cover.
+1. **Follow Core Instructions** — `ostacky.md` Core Instructions section for CodeGraph usage patterns. Use `codegraph_codegraph_explore` to understand the code structure before reading files directly.
+2. **Use CodeGraph for code understanding** — `codegraph_codegraph_explore` on the changed symbols gives you call paths, blast radius, and related code in ONE call. Only `Read` files for details CodeGraph didn't cover.
 3. **Read changed files for full context** — After CodeGraph exploration, read each modified file to understand the surrounding code, imports, types, and control flow.
 4. **Identify the tech stack** — Note languages, frameworks, and libraries in use. This affects what patterns are risky.
 5. **Check for related test files** — For each changed source file, look for corresponding test files. Note whether tests were updated alongside the changes.

package/assets/skills/subagent-driven-development/SKILL.md

@@ -5,14 +5,14 @@ description: Use when Ostacky has confirmed subagent-driven execution mode and d
 
 # Subagent-Driven Development
 
-**IMPORTANT:** Engram is an **MCP server**, not a skill. Tools `mem_save`, `mem_search`, `mem_context` are MCP tools. Do NOT use `skill("engram")` — it doesn't exist.
+**IMPORTANT:** Engram is an **MCP server**, not a skill. Tools `engram_mem_save`, `engram_mem_search`, `engram_mem_context` are MCP tools. Do NOT use `skill("engram")` — it doesn't exist.
 
 **Prerequisite:** This skill runs AFTER the coordinator (Ostacky) has:
 1. Performed execution analysis (via `execution-mode-evaluation` skill)
 2. Presented evidence and recommendation to the user
 3. Received explicit user confirmation for subagent-driven mode
-4. Called `controller.consumeExecutionDecision({ mode: "SUBAGENT_DRIVEN" })`
-5. Checked Engram (`mem_search`) for past subagent execution patterns on similar changes — if a previous approach failed, avoid repeating it
+4. Called `ostacky-controller_consume_execution_decision({ mode: "SUBAGENT_DRIVEN" })`
+5. Checked Engram (`engram_mem_search`) for past subagent execution patterns on similar changes — if a previous approach failed, avoid repeating it
 
 Subagents are execution-only. They never create routing decisions, re-analyze execution mode, or duplicate discovery that the coordinator already resolved.
 
@@ -32,7 +32,7 @@ digraph when_to_use {
     "Tasks mostly independent?" [shape=diamond];
     "Stay in this session?" [shape=diamond];
     "subagent-driven-development" [shape=box];
-    "executing-plans" [shape=box];
+    "Inline execution" [shape=box];
     "Manual execution or brainstorm first" [shape=box];
 
     "Have implementation plan?" -> "Tasks mostly independent?" [label="yes"];
@@ -40,11 +40,11 @@ digraph when_to_use {
     "Tasks mostly independent?" -> "Stay in this session?" [label="yes"];
     "Tasks mostly independent?" -> "Manual execution or brainstorm first" [label="no - tightly coupled"];
     "Stay in this session?" -> "subagent-driven-development" [label="yes"];
-    "Stay in this session?" -> "executing-plans" [label="no - parallel session"];
+    "Stay in this session?" -> "Inline execution" [label="no - parallel session"];
 }
 ```
 
-**vs. Executing Plans (parallel session):**
+**vs. Inline execution (parallel session):**
 - Same session (no context switch)
 - Fresh subagent per task (no context pollution)
 - Two-stage review after each task: spec compliance first, then code quality
@@ -56,7 +56,7 @@ digraph when_to_use {
 
 When constructing the context package for each subagent:
 
-1. Use `codegraph_explore` to get the verbatim source of relevant symbols, their call paths, and blast radius — in ONE call.
+1. Use `codegraph_codegraph_explore` to get the verbatim source of relevant symbols, their call paths, and blast radius — in ONE call.
 2. Only use `Read` for files or details CodeGraph didn't cover.
 3. Include the CodeGraph output in the subagent's context package so they have structural understanding without needing to re-explore.
 
@@ -292,7 +292,7 @@ Done!
 
 **Related workflow skills:**
 - **using-git-worktrees** - Ensures isolated workspace (creates one or verifies existing)
-- **writing-plans** - Creates the plan this skill executes
+- **brainstorming** - Produces design that feeds into implementation (DIRECT) or OpenSpec (SPEC)
 - **review** - Code review template for reviewer subagents
 
 **Subagents should use:**

package/dist/cli.js

@@ -520,31 +520,31 @@ var init_fs = __esm(() => {
 var manifest_default;
 var init_manifest = __esm(() => {
   manifest_default = {
-    version: "0.6.3",
+    version: "0.7.0",
     repo: "JaimeHoracio/Ostacky",
-    tag: "v0.6.3",
+    tag: "v0.7.0",
     agents: [
       {
         name: "ostacky",
         file: "assets/agents/ostacky.md",
-        description: "Orquestador con recuperación automática (nunca se congela), ruteo por nivel de impacto, controller MCP con SDK oficial, edición segura con fallback inline, y delegación en OpenSpec + Superpowers",
-        version: "0.6.3",
-        sha256: "9cd6f45339b897fe002775a3a54f570e47dedd1784fa474d6cd5113594d1bd85"
+        description: "Orquestador con recuperación automática (nunca se congela), ruteo por nivel de impacto, controller MCP con SDK oficial, edición segura con fallback inline, y delegación en OpenSpec + Superpowers. v0.7.0: prune de skills obsoletas, lastHandoff, getAvailableTransitions, consecutiveFailures real, mem_session_summary automático al cierre.",
+        version: "0.7.0",
+        sha256: "f833e69b6eaa3dfe24de9addf6455b3da55f829b51cba3c7c90c65c69cda34c2"
       }
     ],
     commands: [
       {
         name: "install-stack",
         file: "assets/commands/install-stack.md",
-        description: "Instala el stack tecnológico del proyecto (CodeGraph, skills, OpenSpec, Engram, Context7, controller MCP con SDK oficial)",
-        version: "0.6.3",
-        sha256: "6002a342c3570a33c6f5d4bccf767d0c343b06c4a612cfaa8f744ffedd0de34d"
+        description: "Instala el stack tecnológico del proyecto (CodeGraph, skills, OpenSpec, Engram, Context7, controller MCP con SDK oficial). v0.7.0: Paso 1.5 controller expandido con verificación post-instalación y troubleshooting.",
+        version: "0.7.0",
+        sha256: "c957627dfafabfdb9da2717546b996d1d8856f08a6e4c48caf72c562363b566b"
       },
       {
         name: "opsx-sync",
         file: "assets/commands/opsx-sync.md",
         description: "Sincroniza delta specs del change activo sin inicializar CodeGraph si ya existe índice",
-        version: "0.6.3",
+        version: "0.7.0",
         sha256: "fe0158478f2ca63b315037a85fc1632b77868532e319af6c7384285441767d64"
       }
     ],
@@ -552,15 +552,15 @@ var init_manifest = __esm(() => {
       {
         name: "ostacky-controller",
         file: "assets/mcp/ostacky-controller/",
-        description: "Máquina de estados persistida con @modelcontextprotocol/server SDK. 20 tools: start_request → discovery → route → execution → sync → done + validate_edit + complete_task + ping + proceed_to_route + abandon integrado",
-        version: "0.6.3",
-        sha256: "144fb38a88f81a7657131bf416e59bf668c9eca5fd92a607ea94e32a41e1a36d"
+        description: "Máquina de estados persistida con @modelcontextprotocol/server SDK. v0.7.0: 22 tools (incluye set_handoff, get_handoff, clear_handoff, get_available_transitions funcional). Bugfixes: consecutiveFailures real, lastHandoff state, defaultChoice persistido. Robustez: degraded mode automático tras 3 fallos, persistence condicional para Nivel 0, prune de skills obsoletas.",
+        version: "0.7.0",
+        sha256: "0ec6fa0fc2900c78b884258168309653dea4b77ded3596bb56866f689ec2404c"
       },
       {
         name: "openspec",
         file: "assets/mcp/openspec/",
         description: "MCP server local para OpenSpec - proposal, apply, archive, sync de cambios",
-        version: "0.6.3",
+        version: "0.7.0",
         sha256: "fa4be28bfc1e75db8f1a037e2580f04a60066c80e8e5934e90e1041020d04609"
       }
     ],
@@ -568,114 +568,107 @@ var init_manifest = __esm(() => {
       {
         name: "brainstorming",
         file: "assets/skills/brainstorming/SKILL.md",
-        description: "Skill unificado de pensamiento con dos modos: creative-design (producción de diseño → transición a writing-plans o openspec-propose) y open-exploration (exploración libre)",
-        version: "0.6.3",
-        sha256: "602116fb37e5bc879d1a28815a5928c7bcd9563713907ea278669c1cd81fcde1"
+        description: "Skill unificado de pensamiento con dos modos: creative-design (producción de diseño → transición a implementación directa o openspec-propose) y open-exploration (exploración libre)",
+        version: "0.7.0",
+        sha256: "333810ca63450e8ab795e8c5a58e645101e788d79210ca8ac36e10d11be39585"
       },
       {
         name: "execution-mode-evaluation",
         file: "assets/skills/execution-mode-evaluation/SKILL.md",
         description: "Skill de análisis de modo de ejecución — output reconciliado con controller snapshot contract (recommendation field)",
-        version: "0.6.3",
-        sha256: "2e21e69011a0d3cfc2abc4a42b93a616fa9dc7e416955d1f757f6d8e5b35f00b"
-      },
-      {
-        name: "writing-plans",
-        file: "assets/skills/writing-plans/SKILL.md",
-        description: "Skill de planificación de implementación (Superpowers) — execution handoff removido, Ostacky decide modo de ejecución",
-        version: "0.6.3",
-        sha256: "9b14903231df96f54da08158c117801b18bba8f5a2e5a0c069854d059e99f294"
+        version: "0.7.0",
+        sha256: "edf6fca0af52cce83f8fd08e7db9b3d79cc130b53fca505e90d3fbaf4909338b"
       },
       {
         name: "tdd",
         file: "assets/skills/tdd/SKILL.md",
         description: "Skill de test-driven development (Superpowers)",
-        version: "0.6.3",
+        version: "0.7.0",
         sha256: "aa412298980b7826165c211145c1b8e9135f68f36247bb3cb96d0a0eae274486"
       },
       {
         name: "subagent-driven-development",
         file: "assets/skills/subagent-driven-development/SKILL.md",
         description: "Skill de ejecución con subagentes (Superpowers) — ejecuta solo después de confirmación del coordinador Ostacky",
-        version: "0.6.3",
-        sha256: "6128a6cac60eadc11541ba2d37c3e395c6da3a0167767fbb9b7203b1a2a33ebc"
+        version: "0.7.0",
+        sha256: "1a42d714a9a13faf05f0bf7b580e8d837c2e77a5633839542ccce603023419e8"
       },
       {
         name: "dispatching-parallel-agents",
         file: "assets/skills/dispatching-parallel-agents/SKILL.md",
         description: "Skill de dispatch paralelo de agentes (Superpowers)",
-        version: "0.6.3",
+        version: "0.7.0",
         sha256: "3c8a66d51ae2e719e877d02c3dac32bd804722c3bb9227c258ecae78252a20ab"
       },
       {
         name: "review",
         file: "assets/skills/review/SKILL.md",
         description: "Skill de revisión de código (Superpowers)",
-        version: "0.6.3",
-        sha256: "78914222c29306e479d24b09449e1b5a54b5a0482bd5e3e24faace792946a40a"
+        version: "0.7.0",
+        sha256: "b19650ed4d1d4d9857a4dd5b7e08e91328a1a1fc9c45bcc2fb600fa9d0213279"
       },
       {
         name: "receiving-code-review",
         file: "assets/skills/receiving-code-review/SKILL.md",
         description: "Skill de recibir y procesar feedback de code review",
-        version: "0.6.3",
+        version: "0.7.0",
         sha256: "d761e884e71d8d3476ac734d287cae403a4024506a7d12e7361a448217c0a831"
       },
       {
         name: "openspec-propose",
         file: "assets/skills/openspec-propose/SKILL.md",
         description: "Skill de generación de proposal (OpenSpec)",
-        version: "0.6.3",
-        sha256: "a184a9e99c06860e1742a100336b1662b329c5d6b319301449ed0fa01545e1ac"
+        version: "0.7.0",
+        sha256: "bb59100b9fd3c9f9a1ec6509fa975ec55bf7fc737df4fe73bb3e9bbc217b83f7"
       },
       {
         name: "openspec-apply-change",
         file: "assets/skills/openspec-apply-change/SKILL.md",
         description: "Skill de aplicación de change (OpenSpec)",
-        version: "0.6.3",
-        sha256: "b12d80deb4f54f1ceb4792897dbd29fda9b15a71ddd98c30fb6501b085d8eee6"
+        version: "0.7.0",
+        sha256: "dfc823bf89fc7505e91ab6dee9c1f004be38a410bd3d88fb67b211b1b6cbb1d0"
       },
       {
         name: "openspec-archive-change",
         file: "assets/skills/openspec-archive-change/SKILL.md",
         description: "Skill de archivo de change (OpenSpec)",
-        version: "0.6.3",
+        version: "0.7.0",
         sha256: "16e4b561de7747283663fed602e506abf502452c49a8d4b86d7a5539c40f0195"
       },
       {
         name: "openspec-explore",
         file: "assets/skills/openspec-explore/SKILL.md",
         description: "Modo explore para OpenSpec — thinking partner para explorar ideas, investigar problemas y clarificar requisitos antes/durante un cambio",
-        version: "0.6.3",
+        version: "0.7.0",
         sha256: "37ae4aaf17ea71a395bab6dc4d9d61b9a49d7910f6692535ae8553244c46880b"
       },
       {
         name: "using-git-worktrees",
         file: "assets/skills/using-git-worktrees/SKILL.md",
         description: "Skill de uso de git worktrees para aislamiento de trabajo",
-        version: "0.6.3",
+        version: "0.7.0",
         sha256: "11afa9123828fc5caa22137aff907cd99c751e99027fbd4c020751110a949281"
       },
       {
         name: "using-superpowers",
         file: "assets/skills/using-superpowers/SKILL.md",
         description: "Skill de orquestación de Superpowers skills",
-        version: "0.6.3",
+        version: "0.7.0",
         sha256: "7e54536f96d2a379185a10bfc1e970850caa2561b6d8aca7080f0defea0381a4"
       },
       {
         name: "writing-skills",
         file: "assets/skills/writing-skills/SKILL.md",
         description: "Skill de creación y edición de skills",
-        version: "0.6.3",
+        version: "0.7.0",
         sha256: "3d76b906cee518a2b809febb70db95697a35b9f368986bb504b4120c3bfb437a"
       },
       {
         name: "graceful-degradation",
         file: "assets/skills/graceful-degradation/SKILL.md",
         description: "Skill de degradación graceful cuando múltiples tools están indisponibles",
-        version: "0.6.3",
-        sha256: "91c52da5c72a866c5df388fc4d4a3dc317bc55be0040081cc7ae04d3c167a026"
+        version: "0.7.0",
+        sha256: "e5090e475729f21871bb25bce738000f3616955ed3edb3ba926dd740d3298948"
       }
     ]
   };
@@ -1437,7 +1430,7 @@ var init_stack = __esm(() => {
 // package.json
 var package_default = {
   name: "ostacky",
-  version: "0.6.3",
+  version: "0.7.0",
   description: "Instalador interactivo de agentes y comandos para OpenCode",
   type: "module",
   bin: {
@@ -2591,7 +2584,7 @@ async function probeMcpServer(nodeExecutable, serverPath, cwd, statePath) {
         params: {
           protocolVersion: "2025-03-26",
           capabilities: {},
-          clientInfo: { name: "ostacky-installer", version: "0.6.3" }
+          clientInfo: { name: "ostacky-installer", version: "0.7.0" }
         }
       });
     });
@@ -2691,6 +2684,29 @@ async function installSkill(item, manifest, paths) {
   copyDirRecursive(src, dest);
   upsertLockfile(paths, "skills", item, manifest, treeHash);
 }
+function pruneStaleSkills(paths, manifest) {
+  const expected = new Set(manifest.skills.map((s) => s.name));
+  const removed = [];
+  if (!existsSync5(paths.skills))
+    return removed;
+  for (const entry of readdirSync2(paths.skills)) {
+    const dir = join6(paths.skills, entry);
+    try {
+      if (!statSync2(dir).isDirectory())
+        continue;
+    } catch {
+      continue;
+    }
+    if (!expected.has(entry)) {
+      try {
+        rmSync3(dir, { recursive: true, force: true });
+        removeFromLockfile(paths.root, "skills", entry);
+        removed.push(entry);
+      } catch {}
+    }
+  }
+  return removed;
+}
 async function installMcpServer(item, manifest, paths) {
   const src = getBundledMcpPath(item.name);
   if (!existsSync5(src)) {
@@ -3069,6 +3085,10 @@ async function doInstallAll(manifest, paths) {
       errors++;
     }
   }
+  const pruned = pruneStaleSkills(paths, manifest);
+  if (pruned.length > 0) {
+    v2.info(`Skills obsoletas removidas: ${pruned.join(", ")}`);
+  }
   for (const mcp of manifest.mcpServers ?? []) {
     spin.start(`Instalando MCP server: ${mcp.name}  (${mcp.version})`);
     try {
@@ -3282,6 +3302,10 @@ async function doUpdate(manifest, paths) {
     }
   }
   v2.success(`${updated} recurso(s) actualizado(s).`);
+  const pruned = pruneStaleSkills(paths, manifest);
+  if (pruned.length > 0) {
+    v2.info(`Skills obsoletas removidas del filesystem: ${pruned.join(", ")}`);
+  }
 }
 
 // src/prompts/uninstall.ts