@cubis/foundry 0.3.82 → 0.3.83

package/workflows/workflows/agent-environment-setup/platforms/claude/skills/expo-app/SKILL.md

@@ -93,5 +93,12 @@ Guide the design and implementation of production-grade Expo applications using
 - Use `$ARGUMENTS` to access user-provided arguments passed when the skill is invoked.
 - Reference skill-local files with `${CLAUDE_SKILL_DIR}/references/<file>` for portable paths.
 - When `context: fork` is set, the skill runs in an isolated subagent context; the `agent` field names the fork target.
+- Custom subagents in `.claude/agents/*.md` support YAML frontmatter: `name`, `description`, `tools`, `model`, `maxTurns`, `memory`, `handoffs`.
+- Use `model` field in agent frontmatter to select model per subagent (e.g., `model: opus` for complex analysis).
+- Set `maxTurns` to prevent runaway iterations (default: 25, orchestrator: 30).
+- Key agents support `memory: project` for cross-session learning (orchestrator, debugger, researcher, project-planner).
+- Hook templates in `.claude/hooks/` provide lifecycle event integration at `UserPromptSubmit` and other events.
+- Path-scoped rules: `.claude/rules/*.md` with `paths:` frontmatter for targeted guidance.
 - MCP skill tools (`skill_search`, `skill_get`, `skill_validate`, `skill_get_reference`) are available for dynamic skill discovery and loading.
 - Use `allowed-tools` in frontmatter to restrict tool access for security-sensitive skills.
+- Workflow skills can be compiled to `.claude/skills/<workflow-id>/SKILL.md` as `generatedSkills`.

package/workflows/workflows/agent-environment-setup/platforms/claude/skills/express-nodejs/SKILL.md

@@ -80,5 +80,12 @@ Load only what the current step needs.
 - Use `$ARGUMENTS` to access user-provided arguments passed when the skill is invoked.
 - Reference skill-local files with `${CLAUDE_SKILL_DIR}/references/<file>` for portable paths.
 - When `context: fork` is set, the skill runs in an isolated subagent context; the `agent` field names the fork target.
+- Custom subagents in `.claude/agents/*.md` support YAML frontmatter: `name`, `description`, `tools`, `model`, `maxTurns`, `memory`, `handoffs`.
+- Use `model` field in agent frontmatter to select model per subagent (e.g., `model: opus` for complex analysis).
+- Set `maxTurns` to prevent runaway iterations (default: 25, orchestrator: 30).
+- Key agents support `memory: project` for cross-session learning (orchestrator, debugger, researcher, project-planner).
+- Hook templates in `.claude/hooks/` provide lifecycle event integration at `UserPromptSubmit` and other events.
+- Path-scoped rules: `.claude/rules/*.md` with `paths:` frontmatter for targeted guidance.
 - MCP skill tools (`skill_search`, `skill_get`, `skill_validate`, `skill_get_reference`) are available for dynamic skill discovery and loading.
 - Use `allowed-tools` in frontmatter to restrict tool access for security-sensitive skills.
+- Workflow skills can be compiled to `.claude/skills/<workflow-id>/SKILL.md` as `generatedSkills`.

package/workflows/workflows/agent-environment-setup/platforms/claude/skills/fastapi/SKILL.md

@@ -74,5 +74,12 @@ Provide implementation code, Pydantic model definitions, dependency functions, c
 - Use `$ARGUMENTS` to access user-provided arguments passed when the skill is invoked.
 - Reference skill-local files with `${CLAUDE_SKILL_DIR}/references/<file>` for portable paths.
 - When `context: fork` is set, the skill runs in an isolated subagent context; the `agent` field names the fork target.
+- Custom subagents in `.claude/agents/*.md` support YAML frontmatter: `name`, `description`, `tools`, `model`, `maxTurns`, `memory`, `handoffs`.
+- Use `model` field in agent frontmatter to select model per subagent (e.g., `model: opus` for complex analysis).
+- Set `maxTurns` to prevent runaway iterations (default: 25, orchestrator: 30).
+- Key agents support `memory: project` for cross-session learning (orchestrator, debugger, researcher, project-planner).
+- Hook templates in `.claude/hooks/` provide lifecycle event integration at `UserPromptSubmit` and other events.
+- Path-scoped rules: `.claude/rules/*.md` with `paths:` frontmatter for targeted guidance.
 - MCP skill tools (`skill_search`, `skill_get`, `skill_validate`, `skill_get_reference`) are available for dynamic skill discovery and loading.
 - Use `allowed-tools` in frontmatter to restrict tool access for security-sensitive skills.
+- Workflow skills can be compiled to `.claude/skills/<workflow-id>/SKILL.md` as `generatedSkills`.

package/workflows/workflows/agent-environment-setup/platforms/claude/skills/frontend-design/SKILL.md

@@ -93,5 +93,12 @@ Load only what the current task requires.
 - Use `$ARGUMENTS` to access user-provided arguments passed when the skill is invoked.
 - Reference skill-local files with `${CLAUDE_SKILL_DIR}/references/<file>` for portable paths.
 - When `context: fork` is set, the skill runs in an isolated subagent context; the `agent` field names the fork target.
+- Custom subagents in `.claude/agents/*.md` support YAML frontmatter: `name`, `description`, `tools`, `model`, `maxTurns`, `memory`, `handoffs`.
+- Use `model` field in agent frontmatter to select model per subagent (e.g., `model: opus` for complex analysis).
+- Set `maxTurns` to prevent runaway iterations (default: 25, orchestrator: 30).
+- Key agents support `memory: project` for cross-session learning (orchestrator, debugger, researcher, project-planner).
+- Hook templates in `.claude/hooks/` provide lifecycle event integration at `UserPromptSubmit` and other events.
+- Path-scoped rules: `.claude/rules/*.md` with `paths:` frontmatter for targeted guidance.
 - MCP skill tools (`skill_search`, `skill_get`, `skill_validate`, `skill_get_reference`) are available for dynamic skill discovery and loading.
 - Use `allowed-tools` in frontmatter to restrict tool access for security-sensitive skills.
+- Workflow skills can be compiled to `.claude/skills/<workflow-id>/SKILL.md` as `generatedSkills`.

package/workflows/workflows/agent-environment-setup/platforms/claude/skills/gin-golang/SKILL.md

@@ -76,5 +76,12 @@ Load only what the current step needs.
 - Use `$ARGUMENTS` to access user-provided arguments passed when the skill is invoked.
 - Reference skill-local files with `${CLAUDE_SKILL_DIR}/references/<file>` for portable paths.
 - When `context: fork` is set, the skill runs in an isolated subagent context; the `agent` field names the fork target.
+- Custom subagents in `.claude/agents/*.md` support YAML frontmatter: `name`, `description`, `tools`, `model`, `maxTurns`, `memory`, `handoffs`.
+- Use `model` field in agent frontmatter to select model per subagent (e.g., `model: opus` for complex analysis).
+- Set `maxTurns` to prevent runaway iterations (default: 25, orchestrator: 30).
+- Key agents support `memory: project` for cross-session learning (orchestrator, debugger, researcher, project-planner).
+- Hook templates in `.claude/hooks/` provide lifecycle event integration at `UserPromptSubmit` and other events.
+- Path-scoped rules: `.claude/rules/*.md` with `paths:` frontmatter for targeted guidance.
 - MCP skill tools (`skill_search`, `skill_get`, `skill_validate`, `skill_get_reference`) are available for dynamic skill discovery and loading.
 - Use `allowed-tools` in frontmatter to restrict tool access for security-sensitive skills.
+- Workflow skills can be compiled to `.claude/skills/<workflow-id>/SKILL.md` as `generatedSkills`.

package/workflows/workflows/agent-environment-setup/platforms/claude/skills/git-workflow/SKILL.md

@@ -68,5 +68,12 @@ When advising on git workflows, provide:
 - Use `$ARGUMENTS` to access user-provided arguments passed when the skill is invoked.
 - Reference skill-local files with `${CLAUDE_SKILL_DIR}/references/<file>` for portable paths.
 - When `context: fork` is set, the skill runs in an isolated subagent context; the `agent` field names the fork target.
+- Custom subagents in `.claude/agents/*.md` support YAML frontmatter: `name`, `description`, `tools`, `model`, `maxTurns`, `memory`, `handoffs`.
+- Use `model` field in agent frontmatter to select model per subagent (e.g., `model: opus` for complex analysis).
+- Set `maxTurns` to prevent runaway iterations (default: 25, orchestrator: 30).
+- Key agents support `memory: project` for cross-session learning (orchestrator, debugger, researcher, project-planner).
+- Hook templates in `.claude/hooks/` provide lifecycle event integration at `UserPromptSubmit` and other events.
+- Path-scoped rules: `.claude/rules/*.md` with `paths:` frontmatter for targeted guidance.
 - MCP skill tools (`skill_search`, `skill_get`, `skill_validate`, `skill_get_reference`) are available for dynamic skill discovery and loading.
 - Use `allowed-tools` in frontmatter to restrict tool access for security-sensitive skills.
+- Workflow skills can be compiled to `.claude/skills/<workflow-id>/SKILL.md` as `generatedSkills`.

package/workflows/workflows/agent-environment-setup/platforms/claude/skills/go-fiber/SKILL.md

@@ -76,5 +76,12 @@ Load only what the current step needs.
 - Use `$ARGUMENTS` to access user-provided arguments passed when the skill is invoked.
 - Reference skill-local files with `${CLAUDE_SKILL_DIR}/references/<file>` for portable paths.
 - When `context: fork` is set, the skill runs in an isolated subagent context; the `agent` field names the fork target.
+- Custom subagents in `.claude/agents/*.md` support YAML frontmatter: `name`, `description`, `tools`, `model`, `maxTurns`, `memory`, `handoffs`.
+- Use `model` field in agent frontmatter to select model per subagent (e.g., `model: opus` for complex analysis).
+- Set `maxTurns` to prevent runaway iterations (default: 25, orchestrator: 30).
+- Key agents support `memory: project` for cross-session learning (orchestrator, debugger, researcher, project-planner).
+- Hook templates in `.claude/hooks/` provide lifecycle event integration at `UserPromptSubmit` and other events.
+- Path-scoped rules: `.claude/rules/*.md` with `paths:` frontmatter for targeted guidance.
 - MCP skill tools (`skill_search`, `skill_get`, `skill_validate`, `skill_get_reference`) are available for dynamic skill discovery and loading.
 - Use `allowed-tools` in frontmatter to restrict tool access for security-sensitive skills.
+- Workflow skills can be compiled to `.claude/skills/<workflow-id>/SKILL.md` as `generatedSkills`.

package/workflows/workflows/agent-environment-setup/platforms/claude/skills/golang-best-practices/SKILL.md

@@ -79,5 +79,12 @@ No helper scripts are required for this skill right now. Keep execution in `SKIL
 - Use `$ARGUMENTS` to access user-provided arguments passed when the skill is invoked.
 - Reference skill-local files with `${CLAUDE_SKILL_DIR}/references/<file>` for portable paths.
 - When `context: fork` is set, the skill runs in an isolated subagent context; the `agent` field names the fork target.
+- Custom subagents in `.claude/agents/*.md` support YAML frontmatter: `name`, `description`, `tools`, `model`, `maxTurns`, `memory`, `handoffs`.
+- Use `model` field in agent frontmatter to select model per subagent (e.g., `model: opus` for complex analysis).
+- Set `maxTurns` to prevent runaway iterations (default: 25, orchestrator: 30).
+- Key agents support `memory: project` for cross-session learning (orchestrator, debugger, researcher, project-planner).
+- Hook templates in `.claude/hooks/` provide lifecycle event integration at `UserPromptSubmit` and other events.
+- Path-scoped rules: `.claude/rules/*.md` with `paths:` frontmatter for targeted guidance.
 - MCP skill tools (`skill_search`, `skill_get`, `skill_validate`, `skill_get_reference`) are available for dynamic skill discovery and loading.
 - Use `allowed-tools` in frontmatter to restrict tool access for security-sensitive skills.
+- Workflow skills can be compiled to `.claude/skills/<workflow-id>/SKILL.md` as `generatedSkills`.

package/workflows/workflows/agent-environment-setup/platforms/claude/skills/google-workspace/SKILL.md

@@ -91,5 +91,12 @@ Guide the design and implementation of production-grade Google Workspace API int
 - Use `$ARGUMENTS` to access user-provided arguments passed when the skill is invoked.
 - Reference skill-local files with `${CLAUDE_SKILL_DIR}/references/<file>` for portable paths.
 - When `context: fork` is set, the skill runs in an isolated subagent context; the `agent` field names the fork target.
+- Custom subagents in `.claude/agents/*.md` support YAML frontmatter: `name`, `description`, `tools`, `model`, `maxTurns`, `memory`, `handoffs`.
+- Use `model` field in agent frontmatter to select model per subagent (e.g., `model: opus` for complex analysis).
+- Set `maxTurns` to prevent runaway iterations (default: 25, orchestrator: 30).
+- Key agents support `memory: project` for cross-session learning (orchestrator, debugger, researcher, project-planner).
+- Hook templates in `.claude/hooks/` provide lifecycle event integration at `UserPromptSubmit` and other events.
+- Path-scoped rules: `.claude/rules/*.md` with `paths:` frontmatter for targeted guidance.
 - MCP skill tools (`skill_search`, `skill_get`, `skill_validate`, `skill_get_reference`) are available for dynamic skill discovery and loading.
 - Use `allowed-tools` in frontmatter to restrict tool access for security-sensitive skills.
+- Workflow skills can be compiled to `.claude/skills/<workflow-id>/SKILL.md` as `generatedSkills`.

package/workflows/workflows/agent-environment-setup/platforms/claude/skills/huggingface-ml/SKILL.md

@@ -90,5 +90,12 @@ Guide the design and implementation of machine learning workflows using the Hugg
 - Use `$ARGUMENTS` to access user-provided arguments passed when the skill is invoked.
 - Reference skill-local files with `${CLAUDE_SKILL_DIR}/references/<file>` for portable paths.
 - When `context: fork` is set, the skill runs in an isolated subagent context; the `agent` field names the fork target.
+- Custom subagents in `.claude/agents/*.md` support YAML frontmatter: `name`, `description`, `tools`, `model`, `maxTurns`, `memory`, `handoffs`.
+- Use `model` field in agent frontmatter to select model per subagent (e.g., `model: opus` for complex analysis).
+- Set `maxTurns` to prevent runaway iterations (default: 25, orchestrator: 30).
+- Key agents support `memory: project` for cross-session learning (orchestrator, debugger, researcher, project-planner).
+- Hook templates in `.claude/hooks/` provide lifecycle event integration at `UserPromptSubmit` and other events.
+- Path-scoped rules: `.claude/rules/*.md` with `paths:` frontmatter for targeted guidance.
 - MCP skill tools (`skill_search`, `skill_get`, `skill_validate`, `skill_get_reference`) are available for dynamic skill discovery and loading.
 - Use `allowed-tools` in frontmatter to restrict tool access for security-sensitive skills.
+- Workflow skills can be compiled to `.claude/skills/<workflow-id>/SKILL.md` as `generatedSkills`.

package/workflows/workflows/agent-environment-setup/platforms/claude/skills/integration-testing/SKILL.md

@@ -107,5 +107,12 @@ Provide a structured methodology for writing integration tests that verify corre
 - Use `$ARGUMENTS` to access user-provided arguments passed when the skill is invoked.
 - Reference skill-local files with `${CLAUDE_SKILL_DIR}/references/<file>` for portable paths.
 - When `context: fork` is set, the skill runs in an isolated subagent context; the `agent` field names the fork target.
+- Custom subagents in `.claude/agents/*.md` support YAML frontmatter: `name`, `description`, `tools`, `model`, `maxTurns`, `memory`, `handoffs`.
+- Use `model` field in agent frontmatter to select model per subagent (e.g., `model: opus` for complex analysis).
+- Set `maxTurns` to prevent runaway iterations (default: 25, orchestrator: 30).
+- Key agents support `memory: project` for cross-session learning (orchestrator, debugger, researcher, project-planner).
+- Hook templates in `.claude/hooks/` provide lifecycle event integration at `UserPromptSubmit` and other events.
+- Path-scoped rules: `.claude/rules/*.md` with `paths:` frontmatter for targeted guidance.
 - MCP skill tools (`skill_search`, `skill_get`, `skill_validate`, `skill_get_reference`) are available for dynamic skill discovery and loading.
 - Use `allowed-tools` in frontmatter to restrict tool access for security-sensitive skills.
+- Workflow skills can be compiled to `.claude/skills/<workflow-id>/SKILL.md` as `generatedSkills`.

package/workflows/workflows/agent-environment-setup/platforms/claude/skills/java-best-practices/SKILL.md

@@ -74,5 +74,12 @@ Produces Java code using records, sealed types, pattern matching switch expressi
 - Use `$ARGUMENTS` to access user-provided arguments passed when the skill is invoked.
 - Reference skill-local files with `${CLAUDE_SKILL_DIR}/references/<file>` for portable paths.
 - When `context: fork` is set, the skill runs in an isolated subagent context; the `agent` field names the fork target.
+- Custom subagents in `.claude/agents/*.md` support YAML frontmatter: `name`, `description`, `tools`, `model`, `maxTurns`, `memory`, `handoffs`.
+- Use `model` field in agent frontmatter to select model per subagent (e.g., `model: opus` for complex analysis).
+- Set `maxTurns` to prevent runaway iterations (default: 25, orchestrator: 30).
+- Key agents support `memory: project` for cross-session learning (orchestrator, debugger, researcher, project-planner).
+- Hook templates in `.claude/hooks/` provide lifecycle event integration at `UserPromptSubmit` and other events.
+- Path-scoped rules: `.claude/rules/*.md` with `paths:` frontmatter for targeted guidance.
 - MCP skill tools (`skill_search`, `skill_get`, `skill_validate`, `skill_get_reference`) are available for dynamic skill discovery and loading.
 - Use `allowed-tools` in frontmatter to restrict tool access for security-sensitive skills.
+- Workflow skills can be compiled to `.claude/skills/<workflow-id>/SKILL.md` as `generatedSkills`.

package/workflows/workflows/agent-environment-setup/platforms/claude/skills/javascript-best-practices/SKILL.md

@@ -74,5 +74,12 @@ Produces JavaScript code using ESM modules, native runtime APIs, explicit async
 - Use `$ARGUMENTS` to access user-provided arguments passed when the skill is invoked.
 - Reference skill-local files with `${CLAUDE_SKILL_DIR}/references/<file>` for portable paths.
 - When `context: fork` is set, the skill runs in an isolated subagent context; the `agent` field names the fork target.
+- Custom subagents in `.claude/agents/*.md` support YAML frontmatter: `name`, `description`, `tools`, `model`, `maxTurns`, `memory`, `handoffs`.
+- Use `model` field in agent frontmatter to select model per subagent (e.g., `model: opus` for complex analysis).
+- Set `maxTurns` to prevent runaway iterations (default: 25, orchestrator: 30).
+- Key agents support `memory: project` for cross-session learning (orchestrator, debugger, researcher, project-planner).
+- Hook templates in `.claude/hooks/` provide lifecycle event integration at `UserPromptSubmit` and other events.
+- Path-scoped rules: `.claude/rules/*.md` with `paths:` frontmatter for targeted guidance.
 - MCP skill tools (`skill_search`, `skill_get`, `skill_validate`, `skill_get_reference`) are available for dynamic skill discovery and loading.
 - Use `allowed-tools` in frontmatter to restrict tool access for security-sensitive skills.
+- Workflow skills can be compiled to `.claude/skills/<workflow-id>/SKILL.md` as `generatedSkills`.

package/workflows/workflows/agent-environment-setup/platforms/claude/skills/kaizen-iteration/SKILL.md

@@ -100,5 +100,12 @@ Provide a structured, metrics-driven framework for continuous improvement in sof
 - Use `$ARGUMENTS` to access user-provided arguments passed when the skill is invoked.
 - Reference skill-local files with `${CLAUDE_SKILL_DIR}/references/<file>` for portable paths.
 - When `context: fork` is set, the skill runs in an isolated subagent context; the `agent` field names the fork target.
+- Custom subagents in `.claude/agents/*.md` support YAML frontmatter: `name`, `description`, `tools`, `model`, `maxTurns`, `memory`, `handoffs`.
+- Use `model` field in agent frontmatter to select model per subagent (e.g., `model: opus` for complex analysis).
+- Set `maxTurns` to prevent runaway iterations (default: 25, orchestrator: 30).
+- Key agents support `memory: project` for cross-session learning (orchestrator, debugger, researcher, project-planner).
+- Hook templates in `.claude/hooks/` provide lifecycle event integration at `UserPromptSubmit` and other events.
+- Path-scoped rules: `.claude/rules/*.md` with `paths:` frontmatter for targeted guidance.
 - MCP skill tools (`skill_search`, `skill_get`, `skill_validate`, `skill_get_reference`) are available for dynamic skill discovery and loading.
 - Use `allowed-tools` in frontmatter to restrict tool access for security-sensitive skills.
+- Workflow skills can be compiled to `.claude/skills/<workflow-id>/SKILL.md` as `generatedSkills`.

package/workflows/workflows/agent-environment-setup/platforms/claude/skills/kotlin-best-practices/SKILL.md

@@ -74,5 +74,12 @@ Produces Kotlin code using data classes, sealed hierarchies, structured coroutin
 - Use `$ARGUMENTS` to access user-provided arguments passed when the skill is invoked.
 - Reference skill-local files with `${CLAUDE_SKILL_DIR}/references/<file>` for portable paths.
 - When `context: fork` is set, the skill runs in an isolated subagent context; the `agent` field names the fork target.
+- Custom subagents in `.claude/agents/*.md` support YAML frontmatter: `name`, `description`, `tools`, `model`, `maxTurns`, `memory`, `handoffs`.
+- Use `model` field in agent frontmatter to select model per subagent (e.g., `model: opus` for complex analysis).
+- Set `maxTurns` to prevent runaway iterations (default: 25, orchestrator: 30).
+- Key agents support `memory: project` for cross-session learning (orchestrator, debugger, researcher, project-planner).
+- Hook templates in `.claude/hooks/` provide lifecycle event integration at `UserPromptSubmit` and other events.
+- Path-scoped rules: `.claude/rules/*.md` with `paths:` frontmatter for targeted guidance.
 - MCP skill tools (`skill_search`, `skill_get`, `skill_validate`, `skill_get_reference`) are available for dynamic skill discovery and loading.
 - Use `allowed-tools` in frontmatter to restrict tool access for security-sensitive skills.
+- Workflow skills can be compiled to `.claude/skills/<workflow-id>/SKILL.md` as `generatedSkills`.

package/workflows/workflows/agent-environment-setup/platforms/claude/skills/kubernetes-deploy/SKILL.md

@@ -80,5 +80,12 @@ Provide complete Kubernetes manifests in YAML with inline comments explaining no
 - Use `$ARGUMENTS` to access user-provided arguments passed when the skill is invoked.
 - Reference skill-local files with `${CLAUDE_SKILL_DIR}/references/<file>` for portable paths.
 - When `context: fork` is set, the skill runs in an isolated subagent context; the `agent` field names the fork target.
+- Custom subagents in `.claude/agents/*.md` support YAML frontmatter: `name`, `description`, `tools`, `model`, `maxTurns`, `memory`, `handoffs`.
+- Use `model` field in agent frontmatter to select model per subagent (e.g., `model: opus` for complex analysis).
+- Set `maxTurns` to prevent runaway iterations (default: 25, orchestrator: 30).
+- Key agents support `memory: project` for cross-session learning (orchestrator, debugger, researcher, project-planner).
+- Hook templates in `.claude/hooks/` provide lifecycle event integration at `UserPromptSubmit` and other events.
+- Path-scoped rules: `.claude/rules/*.md` with `paths:` frontmatter for targeted guidance.
 - MCP skill tools (`skill_search`, `skill_get`, `skill_validate`, `skill_get_reference`) are available for dynamic skill discovery and loading.
 - Use `allowed-tools` in frontmatter to restrict tool access for security-sensitive skills.
+- Workflow skills can be compiled to `.claude/skills/<workflow-id>/SKILL.md` as `generatedSkills`.

package/workflows/workflows/agent-environment-setup/platforms/claude/skills/laravel/SKILL.md

@@ -68,5 +68,12 @@ No helper scripts are required for this skill right now. Keep execution in `SKIL
 - Use `$ARGUMENTS` to access user-provided arguments passed when the skill is invoked.
 - Reference skill-local files with `${CLAUDE_SKILL_DIR}/references/<file>` for portable paths.
 - When `context: fork` is set, the skill runs in an isolated subagent context; the `agent` field names the fork target.
+- Custom subagents in `.claude/agents/*.md` support YAML frontmatter: `name`, `description`, `tools`, `model`, `maxTurns`, `memory`, `handoffs`.
+- Use `model` field in agent frontmatter to select model per subagent (e.g., `model: opus` for complex analysis).
+- Set `maxTurns` to prevent runaway iterations (default: 25, orchestrator: 30).
+- Key agents support `memory: project` for cross-session learning (orchestrator, debugger, researcher, project-planner).
+- Hook templates in `.claude/hooks/` provide lifecycle event integration at `UserPromptSubmit` and other events.
+- Path-scoped rules: `.claude/rules/*.md` with `paths:` frontmatter for targeted guidance.
 - MCP skill tools (`skill_search`, `skill_get`, `skill_validate`, `skill_get_reference`) are available for dynamic skill discovery and loading.
 - Use `allowed-tools` in frontmatter to restrict tool access for security-sensitive skills.
+- Workflow skills can be compiled to `.claude/skills/<workflow-id>/SKILL.md` as `generatedSkills`.

package/workflows/workflows/agent-environment-setup/platforms/claude/skills/llm-eval/SKILL.md

@@ -74,5 +74,12 @@ Provide eval suite designs as structured JSON or YAML with assertion definitions
 - Use `$ARGUMENTS` to access user-provided arguments passed when the skill is invoked.
 - Reference skill-local files with `${CLAUDE_SKILL_DIR}/references/<file>` for portable paths.
 - When `context: fork` is set, the skill runs in an isolated subagent context; the `agent` field names the fork target.
+- Custom subagents in `.claude/agents/*.md` support YAML frontmatter: `name`, `description`, `tools`, `model`, `maxTurns`, `memory`, `handoffs`.
+- Use `model` field in agent frontmatter to select model per subagent (e.g., `model: opus` for complex analysis).
+- Set `maxTurns` to prevent runaway iterations (default: 25, orchestrator: 30).
+- Key agents support `memory: project` for cross-session learning (orchestrator, debugger, researcher, project-planner).
+- Hook templates in `.claude/hooks/` provide lifecycle event integration at `UserPromptSubmit` and other events.
+- Path-scoped rules: `.claude/rules/*.md` with `paths:` frontmatter for targeted guidance.
 - MCP skill tools (`skill_search`, `skill_get`, `skill_validate`, `skill_get_reference`) are available for dynamic skill discovery and loading.
 - Use `allowed-tools` in frontmatter to restrict tool access for security-sensitive skills.
+- Workflow skills can be compiled to `.claude/skills/<workflow-id>/SKILL.md` as `generatedSkills`.

package/workflows/workflows/agent-environment-setup/platforms/claude/skills/mcp-server-builder/SKILL.md

@@ -92,5 +92,12 @@ Guide the design and implementation of production-grade Model Context Protocol (
 - Use `$ARGUMENTS` to access user-provided arguments passed when the skill is invoked.
 - Reference skill-local files with `${CLAUDE_SKILL_DIR}/references/<file>` for portable paths.
 - When `context: fork` is set, the skill runs in an isolated subagent context; the `agent` field names the fork target.
+- Custom subagents in `.claude/agents/*.md` support YAML frontmatter: `name`, `description`, `tools`, `model`, `maxTurns`, `memory`, `handoffs`.
+- Use `model` field in agent frontmatter to select model per subagent (e.g., `model: opus` for complex analysis).
+- Set `maxTurns` to prevent runaway iterations (default: 25, orchestrator: 30).
+- Key agents support `memory: project` for cross-session learning (orchestrator, debugger, researcher, project-planner).
+- Hook templates in `.claude/hooks/` provide lifecycle event integration at `UserPromptSubmit` and other events.
+- Path-scoped rules: `.claude/rules/*.md` with `paths:` frontmatter for targeted guidance.
 - MCP skill tools (`skill_search`, `skill_get`, `skill_validate`, `skill_get_reference`) are available for dynamic skill discovery and loading.
 - Use `allowed-tools` in frontmatter to restrict tool access for security-sensitive skills.
+- Workflow skills can be compiled to `.claude/skills/<workflow-id>/SKILL.md` as `generatedSkills`.

package/workflows/workflows/agent-environment-setup/platforms/claude/skills/microservices-design/SKILL.md

@@ -84,5 +84,12 @@ Load only what the current task requires.
 - Use `$ARGUMENTS` to access user-provided arguments passed when the skill is invoked.
 - Reference skill-local files with `${CLAUDE_SKILL_DIR}/references/<file>` for portable paths.
 - When `context: fork` is set, the skill runs in an isolated subagent context; the `agent` field names the fork target.
+- Custom subagents in `.claude/agents/*.md` support YAML frontmatter: `name`, `description`, `tools`, `model`, `maxTurns`, `memory`, `handoffs`.
+- Use `model` field in agent frontmatter to select model per subagent (e.g., `model: opus` for complex analysis).
+- Set `maxTurns` to prevent runaway iterations (default: 25, orchestrator: 30).
+- Key agents support `memory: project` for cross-session learning (orchestrator, debugger, researcher, project-planner).
+- Hook templates in `.claude/hooks/` provide lifecycle event integration at `UserPromptSubmit` and other events.
+- Path-scoped rules: `.claude/rules/*.md` with `paths:` frontmatter for targeted guidance.
 - MCP skill tools (`skill_search`, `skill_get`, `skill_validate`, `skill_get_reference`) are available for dynamic skill discovery and loading.
 - Use `allowed-tools` in frontmatter to restrict tool access for security-sensitive skills.
+- Workflow skills can be compiled to `.claude/skills/<workflow-id>/SKILL.md` as `generatedSkills`.

package/workflows/workflows/agent-environment-setup/platforms/claude/skills/nestjs/SKILL.md

@@ -74,5 +74,12 @@ Provide implementation code, module definitions, decorator usage, configuration
 - Use `$ARGUMENTS` to access user-provided arguments passed when the skill is invoked.
 - Reference skill-local files with `${CLAUDE_SKILL_DIR}/references/<file>` for portable paths.
 - When `context: fork` is set, the skill runs in an isolated subagent context; the `agent` field names the fork target.
+- Custom subagents in `.claude/agents/*.md` support YAML frontmatter: `name`, `description`, `tools`, `model`, `maxTurns`, `memory`, `handoffs`.
+- Use `model` field in agent frontmatter to select model per subagent (e.g., `model: opus` for complex analysis).
+- Set `maxTurns` to prevent runaway iterations (default: 25, orchestrator: 30).
+- Key agents support `memory: project` for cross-session learning (orchestrator, debugger, researcher, project-planner).
+- Hook templates in `.claude/hooks/` provide lifecycle event integration at `UserPromptSubmit` and other events.
+- Path-scoped rules: `.claude/rules/*.md` with `paths:` frontmatter for targeted guidance.
 - MCP skill tools (`skill_search`, `skill_get`, `skill_validate`, `skill_get_reference`) are available for dynamic skill discovery and loading.
 - Use `allowed-tools` in frontmatter to restrict tool access for security-sensitive skills.
+- Workflow skills can be compiled to `.claude/skills/<workflow-id>/SKILL.md` as `generatedSkills`.

package/workflows/workflows/agent-environment-setup/platforms/claude/skills/nextjs/SKILL.md

@@ -87,5 +87,12 @@ Load detailed guidance based on context:
 - Use `$ARGUMENTS` to access user-provided arguments passed when the skill is invoked.
 - Reference skill-local files with `${CLAUDE_SKILL_DIR}/references/<file>` for portable paths.
 - When `context: fork` is set, the skill runs in an isolated subagent context; the `agent` field names the fork target.
+- Custom subagents in `.claude/agents/*.md` support YAML frontmatter: `name`, `description`, `tools`, `model`, `maxTurns`, `memory`, `handoffs`.
+- Use `model` field in agent frontmatter to select model per subagent (e.g., `model: opus` for complex analysis).
+- Set `maxTurns` to prevent runaway iterations (default: 25, orchestrator: 30).
+- Key agents support `memory: project` for cross-session learning (orchestrator, debugger, researcher, project-planner).
+- Hook templates in `.claude/hooks/` provide lifecycle event integration at `UserPromptSubmit` and other events.
+- Path-scoped rules: `.claude/rules/*.md` with `paths:` frontmatter for targeted guidance.
 - MCP skill tools (`skill_search`, `skill_get`, `skill_validate`, `skill_get_reference`) are available for dynamic skill discovery and loading.
 - Use `allowed-tools` in frontmatter to restrict tool access for security-sensitive skills.
+- Workflow skills can be compiled to `.claude/skills/<workflow-id>/SKILL.md` as `generatedSkills`.

package/workflows/workflows/agent-environment-setup/platforms/claude/skills/observability/SKILL.md

@@ -78,5 +78,12 @@ Provide complete instrumentation code, configuration files, and alert rule defin
 - Use `$ARGUMENTS` to access user-provided arguments passed when the skill is invoked.
 - Reference skill-local files with `${CLAUDE_SKILL_DIR}/references/<file>` for portable paths.
 - When `context: fork` is set, the skill runs in an isolated subagent context; the `agent` field names the fork target.
+- Custom subagents in `.claude/agents/*.md` support YAML frontmatter: `name`, `description`, `tools`, `model`, `maxTurns`, `memory`, `handoffs`.
+- Use `model` field in agent frontmatter to select model per subagent (e.g., `model: opus` for complex analysis).
+- Set `maxTurns` to prevent runaway iterations (default: 25, orchestrator: 30).
+- Key agents support `memory: project` for cross-session learning (orchestrator, debugger, researcher, project-planner).
+- Hook templates in `.claude/hooks/` provide lifecycle event integration at `UserPromptSubmit` and other events.
+- Path-scoped rules: `.claude/rules/*.md` with `paths:` frontmatter for targeted guidance.
 - MCP skill tools (`skill_search`, `skill_get`, `skill_validate`, `skill_get_reference`) are available for dynamic skill discovery and loading.
 - Use `allowed-tools` in frontmatter to restrict tool access for security-sensitive skills.
+- Workflow skills can be compiled to `.claude/skills/<workflow-id>/SKILL.md` as `generatedSkills`.

package/workflows/workflows/agent-environment-setup/platforms/claude/skills/owasp-security-review/SKILL.md

@@ -105,5 +105,12 @@ Perform structured, OWASP-standards-based security reviews of application code a
 - Use `$ARGUMENTS` to access user-provided arguments passed when the skill is invoked.
 - Reference skill-local files with `${CLAUDE_SKILL_DIR}/references/<file>` for portable paths.
 - When `context: fork` is set, the skill runs in an isolated subagent context; the `agent` field names the fork target.
+- Custom subagents in `.claude/agents/*.md` support YAML frontmatter: `name`, `description`, `tools`, `model`, `maxTurns`, `memory`, `handoffs`.
+- Use `model` field in agent frontmatter to select model per subagent (e.g., `model: opus` for complex analysis).
+- Set `maxTurns` to prevent runaway iterations (default: 25, orchestrator: 30).
+- Key agents support `memory: project` for cross-session learning (orchestrator, debugger, researcher, project-planner).
+- Hook templates in `.claude/hooks/` provide lifecycle event integration at `UserPromptSubmit` and other events.
+- Path-scoped rules: `.claude/rules/*.md` with `paths:` frontmatter for targeted guidance.
 - MCP skill tools (`skill_search`, `skill_get`, `skill_validate`, `skill_get_reference`) are available for dynamic skill discovery and loading.
 - Use `allowed-tools` in frontmatter to restrict tool access for security-sensitive skills.
+- Workflow skills can be compiled to `.claude/skills/<workflow-id>/SKILL.md` as `generatedSkills`.

package/workflows/workflows/agent-environment-setup/platforms/claude/skills/pentest-skill/SKILL.md

@@ -106,5 +106,12 @@ Provide structured guidance for authorized penetration testing engagements follo
 - Use `$ARGUMENTS` to access user-provided arguments passed when the skill is invoked.
 - Reference skill-local files with `${CLAUDE_SKILL_DIR}/references/<file>` for portable paths.
 - When `context: fork` is set, the skill runs in an isolated subagent context; the `agent` field names the fork target.
+- Custom subagents in `.claude/agents/*.md` support YAML frontmatter: `name`, `description`, `tools`, `model`, `maxTurns`, `memory`, `handoffs`.
+- Use `model` field in agent frontmatter to select model per subagent (e.g., `model: opus` for complex analysis).
+- Set `maxTurns` to prevent runaway iterations (default: 25, orchestrator: 30).
+- Key agents support `memory: project` for cross-session learning (orchestrator, debugger, researcher, project-planner).
+- Hook templates in `.claude/hooks/` provide lifecycle event integration at `UserPromptSubmit` and other events.
+- Path-scoped rules: `.claude/rules/*.md` with `paths:` frontmatter for targeted guidance.
 - MCP skill tools (`skill_search`, `skill_get`, `skill_validate`, `skill_get_reference`) are available for dynamic skill discovery and loading.
 - Use `allowed-tools` in frontmatter to restrict tool access for security-sensitive skills.
+- Workflow skills can be compiled to `.claude/skills/<workflow-id>/SKILL.md` as `generatedSkills`.

package/workflows/workflows/agent-environment-setup/platforms/claude/skills/performance-testing/SKILL.md

@@ -120,5 +120,12 @@ Provide a systematic methodology for designing, executing, and interpreting perf
 - Use `$ARGUMENTS` to access user-provided arguments passed when the skill is invoked.
 - Reference skill-local files with `${CLAUDE_SKILL_DIR}/references/<file>` for portable paths.
 - When `context: fork` is set, the skill runs in an isolated subagent context; the `agent` field names the fork target.
+- Custom subagents in `.claude/agents/*.md` support YAML frontmatter: `name`, `description`, `tools`, `model`, `maxTurns`, `memory`, `handoffs`.
+- Use `model` field in agent frontmatter to select model per subagent (e.g., `model: opus` for complex analysis).
+- Set `maxTurns` to prevent runaway iterations (default: 25, orchestrator: 30).
+- Key agents support `memory: project` for cross-session learning (orchestrator, debugger, researcher, project-planner).
+- Hook templates in `.claude/hooks/` provide lifecycle event integration at `UserPromptSubmit` and other events.
+- Path-scoped rules: `.claude/rules/*.md` with `paths:` frontmatter for targeted guidance.
 - MCP skill tools (`skill_search`, `skill_get`, `skill_validate`, `skill_get_reference`) are available for dynamic skill discovery and loading.
 - Use `allowed-tools` in frontmatter to restrict tool access for security-sensitive skills.
+- Workflow skills can be compiled to `.claude/skills/<workflow-id>/SKILL.md` as `generatedSkills`.

package/workflows/workflows/agent-environment-setup/platforms/claude/skills/php-best-practices/SKILL.md

@@ -87,5 +87,12 @@ No helper scripts are required for this skill right now. Keep execution in `SKIL
 - Use `$ARGUMENTS` to access user-provided arguments passed when the skill is invoked.
 - Reference skill-local files with `${CLAUDE_SKILL_DIR}/references/<file>` for portable paths.
 - When `context: fork` is set, the skill runs in an isolated subagent context; the `agent` field names the fork target.
+- Custom subagents in `.claude/agents/*.md` support YAML frontmatter: `name`, `description`, `tools`, `model`, `maxTurns`, `memory`, `handoffs`.
+- Use `model` field in agent frontmatter to select model per subagent (e.g., `model: opus` for complex analysis).
+- Set `maxTurns` to prevent runaway iterations (default: 25, orchestrator: 30).
+- Key agents support `memory: project` for cross-session learning (orchestrator, debugger, researcher, project-planner).
+- Hook templates in `.claude/hooks/` provide lifecycle event integration at `UserPromptSubmit` and other events.
+- Path-scoped rules: `.claude/rules/*.md` with `paths:` frontmatter for targeted guidance.
 - MCP skill tools (`skill_search`, `skill_get`, `skill_validate`, `skill_get_reference`) are available for dynamic skill discovery and loading.
 - Use `allowed-tools` in frontmatter to restrict tool access for security-sensitive skills.
+- Workflow skills can be compiled to `.claude/skills/<workflow-id>/SKILL.md` as `generatedSkills`.

package/workflows/workflows/agent-environment-setup/platforms/claude/skills/playwright-interactive/SKILL.md

@@ -85,5 +85,12 @@ Deliver:
 - Use `$ARGUMENTS` to access user-provided arguments passed when the skill is invoked.
 - Reference skill-local files with `${CLAUDE_SKILL_DIR}/references/<file>` for portable paths.
 - When `context: fork` is set, the skill runs in an isolated subagent context; the `agent` field names the fork target.
+- Custom subagents in `.claude/agents/*.md` support YAML frontmatter: `name`, `description`, `tools`, `model`, `maxTurns`, `memory`, `handoffs`.
+- Use `model` field in agent frontmatter to select model per subagent (e.g., `model: opus` for complex analysis).
+- Set `maxTurns` to prevent runaway iterations (default: 25, orchestrator: 30).
+- Key agents support `memory: project` for cross-session learning (orchestrator, debugger, researcher, project-planner).
+- Hook templates in `.claude/hooks/` provide lifecycle event integration at `UserPromptSubmit` and other events.
+- Path-scoped rules: `.claude/rules/*.md` with `paths:` frontmatter for targeted guidance.
 - MCP skill tools (`skill_search`, `skill_get`, `skill_validate`, `skill_get_reference`) are available for dynamic skill discovery and loading.
 - Use `allowed-tools` in frontmatter to restrict tool access for security-sensitive skills.
+- Workflow skills can be compiled to `.claude/skills/<workflow-id>/SKILL.md` as `generatedSkills`.

package/workflows/workflows/agent-environment-setup/platforms/claude/skills/playwright-persistent-browser/SKILL.md

@@ -81,5 +81,12 @@ Deliver:
 - Use `$ARGUMENTS` to access user-provided arguments passed when the skill is invoked.
 - Reference skill-local files with `${CLAUDE_SKILL_DIR}/references/<file>` for portable paths.
 - When `context: fork` is set, the skill runs in an isolated subagent context; the `agent` field names the fork target.
+- Custom subagents in `.claude/agents/*.md` support YAML frontmatter: `name`, `description`, `tools`, `model`, `maxTurns`, `memory`, `handoffs`.
+- Use `model` field in agent frontmatter to select model per subagent (e.g., `model: opus` for complex analysis).
+- Set `maxTurns` to prevent runaway iterations (default: 25, orchestrator: 30).
+- Key agents support `memory: project` for cross-session learning (orchestrator, debugger, researcher, project-planner).
+- Hook templates in `.claude/hooks/` provide lifecycle event integration at `UserPromptSubmit` and other events.
+- Path-scoped rules: `.claude/rules/*.md` with `paths:` frontmatter for targeted guidance.
 - MCP skill tools (`skill_search`, `skill_get`, `skill_validate`, `skill_get_reference`) are available for dynamic skill discovery and loading.
 - Use `allowed-tools` in frontmatter to restrict tool access for security-sensitive skills.
+- Workflow skills can be compiled to `.claude/skills/<workflow-id>/SKILL.md` as `generatedSkills`.

package/workflows/workflows/agent-environment-setup/platforms/claude/skills/prisma/SKILL.md

@@ -77,5 +77,12 @@ Load only what the current task requires.
 - Use `$ARGUMENTS` to access user-provided arguments passed when the skill is invoked.
 - Reference skill-local files with `${CLAUDE_SKILL_DIR}/references/<file>` for portable paths.
 - When `context: fork` is set, the skill runs in an isolated subagent context; the `agent` field names the fork target.
+- Custom subagents in `.claude/agents/*.md` support YAML frontmatter: `name`, `description`, `tools`, `model`, `maxTurns`, `memory`, `handoffs`.
+- Use `model` field in agent frontmatter to select model per subagent (e.g., `model: opus` for complex analysis).
+- Set `maxTurns` to prevent runaway iterations (default: 25, orchestrator: 30).
+- Key agents support `memory: project` for cross-session learning (orchestrator, debugger, researcher, project-planner).
+- Hook templates in `.claude/hooks/` provide lifecycle event integration at `UserPromptSubmit` and other events.
+- Path-scoped rules: `.claude/rules/*.md` with `paths:` frontmatter for targeted guidance.
 - MCP skill tools (`skill_search`, `skill_get`, `skill_validate`, `skill_get_reference`) are available for dynamic skill discovery and loading.
 - Use `allowed-tools` in frontmatter to restrict tool access for security-sensitive skills.
+- Workflow skills can be compiled to `.claude/skills/<workflow-id>/SKILL.md` as `generatedSkills`.

package/workflows/workflows/agent-environment-setup/platforms/claude/skills/prompt-engineering/SKILL.md

@@ -80,5 +80,12 @@ Provide system prompts as formatted text blocks with labeled sections, few-shot
 - Use `$ARGUMENTS` to access user-provided arguments passed when the skill is invoked.
 - Reference skill-local files with `${CLAUDE_SKILL_DIR}/references/<file>` for portable paths.
 - When `context: fork` is set, the skill runs in an isolated subagent context; the `agent` field names the fork target.
+- Custom subagents in `.claude/agents/*.md` support YAML frontmatter: `name`, `description`, `tools`, `model`, `maxTurns`, `memory`, `handoffs`.
+- Use `model` field in agent frontmatter to select model per subagent (e.g., `model: opus` for complex analysis).
+- Set `maxTurns` to prevent runaway iterations (default: 25, orchestrator: 30).
+- Key agents support `memory: project` for cross-session learning (orchestrator, debugger, researcher, project-planner).
+- Hook templates in `.claude/hooks/` provide lifecycle event integration at `UserPromptSubmit` and other events.
+- Path-scoped rules: `.claude/rules/*.md` with `paths:` frontmatter for targeted guidance.
 - MCP skill tools (`skill_search`, `skill_get`, `skill_validate`, `skill_get_reference`) are available for dynamic skill discovery and loading.
 - Use `allowed-tools` in frontmatter to restrict tool access for security-sensitive skills.
+- Workflow skills can be compiled to `.claude/skills/<workflow-id>/SKILL.md` as `generatedSkills`.

package/workflows/workflows/agent-environment-setup/platforms/claude/skills/python-best-practices/SKILL.md

@@ -83,5 +83,12 @@ No helper scripts are required for this skill right now. Keep execution in `SKIL
 - Use `$ARGUMENTS` to access user-provided arguments passed when the skill is invoked.
 - Reference skill-local files with `${CLAUDE_SKILL_DIR}/references/<file>` for portable paths.
 - When `context: fork` is set, the skill runs in an isolated subagent context; the `agent` field names the fork target.
+- Custom subagents in `.claude/agents/*.md` support YAML frontmatter: `name`, `description`, `tools`, `model`, `maxTurns`, `memory`, `handoffs`.
+- Use `model` field in agent frontmatter to select model per subagent (e.g., `model: opus` for complex analysis).
+- Set `maxTurns` to prevent runaway iterations (default: 25, orchestrator: 30).
+- Key agents support `memory: project` for cross-session learning (orchestrator, debugger, researcher, project-planner).
+- Hook templates in `.claude/hooks/` provide lifecycle event integration at `UserPromptSubmit` and other events.
+- Path-scoped rules: `.claude/rules/*.md` with `paths:` frontmatter for targeted guidance.
 - MCP skill tools (`skill_search`, `skill_get`, `skill_validate`, `skill_get_reference`) are available for dynamic skill discovery and loading.
 - Use `allowed-tools` in frontmatter to restrict tool access for security-sensitive skills.
+- Workflow skills can be compiled to `.claude/skills/<workflow-id>/SKILL.md` as `generatedSkills`.

package/workflows/workflows/agent-environment-setup/platforms/claude/skills/rag-patterns/SKILL.md

@@ -75,5 +75,12 @@ Provide architecture diagrams as ASCII or Mermaid, configuration snippets for ve
 - Use `$ARGUMENTS` to access user-provided arguments passed when the skill is invoked.
 - Reference skill-local files with `${CLAUDE_SKILL_DIR}/references/<file>` for portable paths.
 - When `context: fork` is set, the skill runs in an isolated subagent context; the `agent` field names the fork target.
+- Custom subagents in `.claude/agents/*.md` support YAML frontmatter: `name`, `description`, `tools`, `model`, `maxTurns`, `memory`, `handoffs`.
+- Use `model` field in agent frontmatter to select model per subagent (e.g., `model: opus` for complex analysis).
+- Set `maxTurns` to prevent runaway iterations (default: 25, orchestrator: 30).
+- Key agents support `memory: project` for cross-session learning (orchestrator, debugger, researcher, project-planner).
+- Hook templates in `.claude/hooks/` provide lifecycle event integration at `UserPromptSubmit` and other events.
+- Path-scoped rules: `.claude/rules/*.md` with `paths:` frontmatter for targeted guidance.
 - MCP skill tools (`skill_search`, `skill_get`, `skill_validate`, `skill_get_reference`) are available for dynamic skill discovery and loading.
 - Use `allowed-tools` in frontmatter to restrict tool access for security-sensitive skills.
+- Workflow skills can be compiled to `.claude/skills/<workflow-id>/SKILL.md` as `generatedSkills`.

package/workflows/workflows/agent-environment-setup/platforms/claude/skills/react/SKILL.md

@@ -87,5 +87,12 @@ Load detailed guidance based on context:
 - Use `$ARGUMENTS` to access user-provided arguments passed when the skill is invoked.
 - Reference skill-local files with `${CLAUDE_SKILL_DIR}/references/<file>` for portable paths.
 - When `context: fork` is set, the skill runs in an isolated subagent context; the `agent` field names the fork target.
+- Custom subagents in `.claude/agents/*.md` support YAML frontmatter: `name`, `description`, `tools`, `model`, `maxTurns`, `memory`, `handoffs`.
+- Use `model` field in agent frontmatter to select model per subagent (e.g., `model: opus` for complex analysis).
+- Set `maxTurns` to prevent runaway iterations (default: 25, orchestrator: 30).
+- Key agents support `memory: project` for cross-session learning (orchestrator, debugger, researcher, project-planner).
+- Hook templates in `.claude/hooks/` provide lifecycle event integration at `UserPromptSubmit` and other events.
+- Path-scoped rules: `.claude/rules/*.md` with `paths:` frontmatter for targeted guidance.
 - MCP skill tools (`skill_search`, `skill_get`, `skill_validate`, `skill_get_reference`) are available for dynamic skill discovery and loading.
 - Use `allowed-tools` in frontmatter to restrict tool access for security-sensitive skills.
+- Workflow skills can be compiled to `.claude/skills/<workflow-id>/SKILL.md` as `generatedSkills`.

package/workflows/workflows/agent-environment-setup/platforms/claude/skills/react-native/SKILL.md

@@ -86,5 +86,12 @@ Load detailed guidance based on context:
 - Use `$ARGUMENTS` to access user-provided arguments passed when the skill is invoked.
 - Reference skill-local files with `${CLAUDE_SKILL_DIR}/references/<file>` for portable paths.
 - When `context: fork` is set, the skill runs in an isolated subagent context; the `agent` field names the fork target.
+- Custom subagents in `.claude/agents/*.md` support YAML frontmatter: `name`, `description`, `tools`, `model`, `maxTurns`, `memory`, `handoffs`.
+- Use `model` field in agent frontmatter to select model per subagent (e.g., `model: opus` for complex analysis).
+- Set `maxTurns` to prevent runaway iterations (default: 25, orchestrator: 30).
+- Key agents support `memory: project` for cross-session learning (orchestrator, debugger, researcher, project-planner).
+- Hook templates in `.claude/hooks/` provide lifecycle event integration at `UserPromptSubmit` and other events.
+- Path-scoped rules: `.claude/rules/*.md` with `paths:` frontmatter for targeted guidance.
 - MCP skill tools (`skill_search`, `skill_get`, `skill_validate`, `skill_get_reference`) are available for dynamic skill discovery and loading.
 - Use `allowed-tools` in frontmatter to restrict tool access for security-sensitive skills.
+- Workflow skills can be compiled to `.claude/skills/<workflow-id>/SKILL.md` as `generatedSkills`.

package/workflows/workflows/agent-environment-setup/platforms/claude/skills/react-native-callstack/SKILL.md

@@ -90,5 +90,12 @@ Guide the design and implementation of high-performance React Native application
 - Use `$ARGUMENTS` to access user-provided arguments passed when the skill is invoked.
 - Reference skill-local files with `${CLAUDE_SKILL_DIR}/references/<file>` for portable paths.
 - When `context: fork` is set, the skill runs in an isolated subagent context; the `agent` field names the fork target.
+- Custom subagents in `.claude/agents/*.md` support YAML frontmatter: `name`, `description`, `tools`, `model`, `maxTurns`, `memory`, `handoffs`.
+- Use `model` field in agent frontmatter to select model per subagent (e.g., `model: opus` for complex analysis).
+- Set `maxTurns` to prevent runaway iterations (default: 25, orchestrator: 30).
+- Key agents support `memory: project` for cross-session learning (orchestrator, debugger, researcher, project-planner).
+- Hook templates in `.claude/hooks/` provide lifecycle event integration at `UserPromptSubmit` and other events.
+- Path-scoped rules: `.claude/rules/*.md` with `paths:` frontmatter for targeted guidance.
 - MCP skill tools (`skill_search`, `skill_get`, `skill_validate`, `skill_get_reference`) are available for dynamic skill discovery and loading.
 - Use `allowed-tools` in frontmatter to restrict tool access for security-sensitive skills.
+- Workflow skills can be compiled to `.claude/skills/<workflow-id>/SKILL.md` as `generatedSkills`.

package/workflows/workflows/agent-environment-setup/platforms/claude/skills/receiving-code-review/SKILL.md

@@ -74,5 +74,12 @@ When responding to review feedback, provide:
 - Use `$ARGUMENTS` to access user-provided arguments passed when the skill is invoked.
 - Reference skill-local files with `${CLAUDE_SKILL_DIR}/references/<file>` for portable paths.
 - When `context: fork` is set, the skill runs in an isolated subagent context; the `agent` field names the fork target.
+- Custom subagents in `.claude/agents/*.md` support YAML frontmatter: `name`, `description`, `tools`, `model`, `maxTurns`, `memory`, `handoffs`.
+- Use `model` field in agent frontmatter to select model per subagent (e.g., `model: opus` for complex analysis).
+- Set `maxTurns` to prevent runaway iterations (default: 25, orchestrator: 30).
+- Key agents support `memory: project` for cross-session learning (orchestrator, debugger, researcher, project-planner).
+- Hook templates in `.claude/hooks/` provide lifecycle event integration at `UserPromptSubmit` and other events.
+- Path-scoped rules: `.claude/rules/*.md` with `paths:` frontmatter for targeted guidance.
 - MCP skill tools (`skill_search`, `skill_get`, `skill_validate`, `skill_get_reference`) are available for dynamic skill discovery and loading.
 - Use `allowed-tools` in frontmatter to restrict tool access for security-sensitive skills.
+- Workflow skills can be compiled to `.claude/skills/<workflow-id>/SKILL.md` as `generatedSkills`.

package/workflows/workflows/agent-environment-setup/platforms/claude/skills/remix/SKILL.md

@@ -78,5 +78,12 @@ Load only what the current task requires.
 - Use `$ARGUMENTS` to access user-provided arguments passed when the skill is invoked.
 - Reference skill-local files with `${CLAUDE_SKILL_DIR}/references/<file>` for portable paths.
 - When `context: fork` is set, the skill runs in an isolated subagent context; the `agent` field names the fork target.
+- Custom subagents in `.claude/agents/*.md` support YAML frontmatter: `name`, `description`, `tools`, `model`, `maxTurns`, `memory`, `handoffs`.
+- Use `model` field in agent frontmatter to select model per subagent (e.g., `model: opus` for complex analysis).
+- Set `maxTurns` to prevent runaway iterations (default: 25, orchestrator: 30).
+- Key agents support `memory: project` for cross-session learning (orchestrator, debugger, researcher, project-planner).
+- Hook templates in `.claude/hooks/` provide lifecycle event integration at `UserPromptSubmit` and other events.
+- Path-scoped rules: `.claude/rules/*.md` with `paths:` frontmatter for targeted guidance.
 - MCP skill tools (`skill_search`, `skill_get`, `skill_validate`, `skill_get_reference`) are available for dynamic skill discovery and loading.
 - Use `allowed-tools` in frontmatter to restrict tool access for security-sensitive skills.
+- Workflow skills can be compiled to `.claude/skills/<workflow-id>/SKILL.md` as `generatedSkills`.

package/workflows/workflows/agent-environment-setup/platforms/claude/skills/requesting-code-review/SKILL.md

@@ -75,5 +75,12 @@ When preparing a review request, provide:
 - Use `$ARGUMENTS` to access user-provided arguments passed when the skill is invoked.
 - Reference skill-local files with `${CLAUDE_SKILL_DIR}/references/<file>` for portable paths.
 - When `context: fork` is set, the skill runs in an isolated subagent context; the `agent` field names the fork target.
+- Custom subagents in `.claude/agents/*.md` support YAML frontmatter: `name`, `description`, `tools`, `model`, `maxTurns`, `memory`, `handoffs`.
+- Use `model` field in agent frontmatter to select model per subagent (e.g., `model: opus` for complex analysis).
+- Set `maxTurns` to prevent runaway iterations (default: 25, orchestrator: 30).
+- Key agents support `memory: project` for cross-session learning (orchestrator, debugger, researcher, project-planner).
+- Hook templates in `.claude/hooks/` provide lifecycle event integration at `UserPromptSubmit` and other events.
+- Path-scoped rules: `.claude/rules/*.md` with `paths:` frontmatter for targeted guidance.
 - MCP skill tools (`skill_search`, `skill_get`, `skill_validate`, `skill_get_reference`) are available for dynamic skill discovery and loading.
 - Use `allowed-tools` in frontmatter to restrict tool access for security-sensitive skills.
+- Workflow skills can be compiled to `.claude/skills/<workflow-id>/SKILL.md` as `generatedSkills`.