tech-lead-stack 1.0.1

package/.agents/hr-workflows/hr-ad-distributor.md

@@ -0,0 +1,18 @@
+---
+name: hr-ad-distributor
+description: Multi-Channel Job Ad Distributor
+modes:
+  - write
+---
+
+// turbo
+
+**MANDATORY: You MUST call the get_skill tool (which may be prefixed as mcp_tech-lead-stack_get_skill or tech-lead-stack_get_skill) with skillName: "hr-$1" before proceeding.**
+
+1. **Phase 1: Skill Acquisition**: Call the `get_skill` tool:
+   - skillName: "hr-ad-distributor"
+   - projectName: "<PROJECT_NAME>"
+   - model: "Gemini"
+   - agent: "HR-Assistant"
+
+2. **Phase 2: Distribution**: Follow the acquired skill's workflow to prepare a parity-checked distribution manifest for the approved JD across LinkedIn and Ashby.

package/.agents/hr-workflows/hr-candidate-sourcer.md

@@ -0,0 +1,18 @@
+---
+name: hr-candidate-sourcer
+description: Passive Candidate Sourcer
+modes:
+  - write
+---
+
+// turbo
+
+**MANDATORY: You MUST call the get_skill tool (which may be prefixed as mcp_tech-lead-stack_get_skill or tech-lead-stack_get_skill) with skillName: "hr-$1" before proceeding.**
+
+1. **Phase 1: Skill Acquisition**: Call the `get_skill` tool:
+   - skillName: "hr-candidate-sourcer"
+   - projectName: "<PROJECT_NAME>"
+   - model: "Gemini"
+   - agent: "HR-Assistant"
+
+2. **Phase 2: Sourcing**: Follow the acquired skill's workflow to produce a compliance-checked, match-scored sourcing longlist of passive candidates.

package/.agents/hr-workflows/hr-endorsement-synthesizer.md

@@ -0,0 +1,18 @@
+---
+name: hr-endorsement-synthesizer
+description: Evidence-Traceable Endorsement Synthesizer
+modes:
+  - write
+---
+
+// turbo
+
+**MANDATORY: You MUST call the get_skill tool (which may be prefixed as mcp_tech-lead-stack_get_skill or tech-lead-stack_get_skill) with skillName: "hr-$1" before proceeding.**
+
+1. **Phase 1: Skill Acquisition**: Call the `get_skill` tool:
+   - skillName: "hr-endorsement-synthesizer"
+   - projectName: "<PROJECT_NAME>"
+   - model: "Gemini"
+   - agent: "HR-Assistant"
+
+2. **Phase 2: Endorsement Synthesis**: Follow the acquired skill's workflow to synthesize resumes and interview notes into an evidence-traceable endorsement report of the top 3-5 applicants.

package/.agents/hr-workflows/hr-intake-specifier.md

@@ -0,0 +1,18 @@
+---
+name: hr-intake-specifier
+description: Requisition Intake Brief Specifier
+modes:
+  - write
+---
+
+// turbo
+
+**MANDATORY: You MUST call the get_skill tool (which may be prefixed as mcp_tech-lead-stack_get_skill or tech-lead-stack_get_skill) with skillName: "hr-$1" before proceeding.**
+
+1. **Phase 1: Skill Acquisition**: Call the `get_skill` tool:
+   - skillName: "hr-intake-specifier"
+   - projectName: "<PROJECT_NAME>"
+   - model: "Gemini"
+   - agent: "HR-Assistant"
+
+2. **Phase 2: Intake Drafting**: Follow the acquired skill's workflow to translate the client kickoff call into a complete, deduplicated requisition intake brief.

package/.agents/hr-workflows/hr-interview-auditor.md

@@ -0,0 +1,18 @@
+---
+name: hr-interview-auditor
+description: Structured Interview Evaluator
+modes:
+  - write
+---
+
+// turbo
+
+**MANDATORY: You MUST call the get_skill tool (which may be prefixed as mcp_tech-lead-stack_get_skill or tech-lead-stack_get_skill) with skillName: "hr-$1" before proceeding.**
+
+1. **Phase 1: Skill Acquisition**: Call the `get_skill` tool:
+   - skillName: "hr-interview-auditor"
+   - projectName: "<PROJECT_NAME>"
+   - model: "Gemini"
+   - agent: "HR-Assistant"
+
+2. **Phase 2: Interview Evaluation**: Follow the acquired skill's workflow to evaluate the applicant against the requisition scorecard and capture evidence-backed interview notes.

package/.agents/hr-workflows/hr-jd-drafter.md

@@ -0,0 +1,18 @@
+---
+name: hr-jd-drafter
+description: Inclusive Job Description Drafter
+modes:
+  - write
+---
+
+// turbo
+
+**MANDATORY: You MUST call the get_skill tool (which may be prefixed as mcp_tech-lead-stack_get_skill or tech-lead-stack_get_skill) with skillName: "hr-$1" before proceeding.**
+
+1. **Phase 1: Skill Acquisition**: Call the `get_skill` tool:
+   - skillName: "hr-jd-drafter"
+   - projectName: "<PROJECT_NAME>"
+   - model: "Gemini"
+   - agent: "HR-Assistant"
+
+2. **Phase 2: JD Drafting**: Follow the acquired skill's workflow to produce a finalized, inclusive, market-ready Job Description from the intake brief or client-provided draft.

package/.agents/hr-workflows/hr-pipeline-translator.md

@@ -0,0 +1,18 @@
+---
+name: hr-pipeline-translator
+description: Client-Facing Pipeline Status Translator
+modes:
+  - write
+---
+
+// turbo
+
+**MANDATORY: You MUST call the get_skill tool (which may be prefixed as mcp_tech-lead-stack_get_skill or tech-lead-stack_get_skill) with skillName: "hr-$1" before proceeding.**
+
+1. **Phase 1: Skill Acquisition**: Call the `get_skill` tool:
+   - skillName: "hr-pipeline-translator"
+   - projectName: "<PROJECT_NAME>"
+   - model: "Gemini"
+   - agent: "HR-Assistant"
+
+2. **Phase 2: Update Translation**: Follow the acquired skill's workflow to translate recent ATS pipeline movement into a clear, client-facing recruitment status update.

package/.agents/pm-workflows/pm-action-item-mapper.md

@@ -0,0 +1,18 @@
+---
+name: pm-action-item-mapper
+description: Meeting Note Technical Action-Item Mapper
+modes:
+  - write
+---
+
+// turbo
+
+**MANDATORY: You MUST call the get_skill tool (which may be prefixed as mcp_tech-lead-stack_get_skill or tech-lead-stack_get_skill) with skillName: "pm-$1" before proceeding.**
+
+1. **Phase 1: Skill Acquisition**: Call the `get_skill` tool:
+   - skillName: "pm-action-item-mapper"
+   - projectName: "<PROJECT_NAME>"
+   - model: "Gemini"
+   - agent: "PM-Assistant"
+
+2. **Phase 2: Item Mapping**: Follow the acquired skill's workflow to translate meeting decisions into actionable technical tasks linked to specific parts of the codebase.

package/.agents/pm-workflows/pm-backlog-auditor.md

@@ -0,0 +1,18 @@
+---
+name: pm-backlog-auditor
+description: Backlog Sufficiency & Logic Auditor
+modes:
+  - write
+---
+
+// turbo
+
+**MANDATORY: You MUST call the get_skill tool (which may be prefixed as mcp_tech-lead-stack_get_skill or tech-lead-stack_get_skill) with skillName: "pm-$1" before proceeding.**
+
+1. **Phase 1: Skill Acquisition**: Call the `get_skill` tool:
+   - skillName: "pm-backlog-auditor"
+   - projectName: "<PROJECT_NAME>"
+   - model: "Gemini"
+   - agent: "PM-Assistant"
+
+2. **Phase 2: Logic Audit**: Follow the acquired skill's workflow to validate the backlog for consistency, feasibility, and technical logic.

package/.agents/pm-workflows/pm-context-summarizer.md

@@ -0,0 +1,18 @@
+---
+name: pm-context-summarizer
+description: Pre-Meeting Development Context Summarizer
+modes:
+  - write
+---
+
+// turbo
+
+**MANDATORY: You MUST call the get_skill tool (which may be prefixed as mcp_tech-lead-stack_get_skill or tech-lead-stack_get_skill) with skillName: "pm-$1" before proceeding.**
+
+1. **Phase 1: Skill Acquisition**: Call the `get_skill` tool:
+   - skillName: "pm-context-summarizer"
+   - projectName: "<PROJECT_NAME>"
+   - model: "Gemini"
+   - agent: "PM-Assistant"
+
+2. **Phase 2: Context Synthesis**: Follow the acquired skill's workflow to summarize recent git activity and blockers into a non-technical meeting briefing.

package/.agents/pm-workflows/pm-design-system-auditor.md

@@ -0,0 +1,18 @@
+---
+name: pm-design-system-auditor
+description: Design System Implementation Auditor
+modes:
+  - write
+---
+
+// turbo
+
+**MANDATORY: You MUST call the get_skill tool (which may be prefixed as mcp_tech-lead-stack_get_skill or tech-lead-stack_get_skill) with skillName: "pm-$1" before proceeding.**
+
+1. **Phase 1: Skill Acquisition**: Call the `get_skill` tool:
+   - skillName: "design-system-review"
+   - projectName: "<PROJECT_NAME>"
+   - model: "Gemini"
+   - agent: "PM-Assistant"
+
+2. **Phase 2: Consistency Audit**: Follow the acquired skill's workflow to audit the UI implementation against design system tokens and patterns.

package/.agents/pm-workflows/pm-effort-estimator.md

@@ -0,0 +1,18 @@
+---
+name: pm-effort-estimator
+description: Historical Complexity & Effort Estimator
+modes:
+  - write
+---
+
+// turbo
+
+**MANDATORY: You MUST call the get_skill tool (which may be prefixed as mcp_tech-lead-stack_get_skill or tech-lead-stack_get_skill) with skillName: "pm-$1" before proceeding.**
+
+1. **Phase 1: Skill Acquisition**: Call the `get_skill` tool:
+   - skillName: "pm-effort-estimator"
+   - projectName: "<PROJECT_NAME>"
+   - model: "Gemini"
+   - agent: "PM-Assistant"
+
+2. **Phase 2: Estimate Generation**: Follow the acquired skill's workflow to provide a data-driven effort estimation based on the codebase's historical churn and complexity.

package/.agents/pm-workflows/pm-newsletter-generator.md

@@ -0,0 +1,18 @@
+---
+name: pm-newsletter-generator
+description: Commit-to-Newsletter Content Summarizer
+modes:
+  - write
+---
+
+// turbo
+
+**MANDATORY: You MUST call the get_skill tool (which may be prefixed as mcp_tech-lead-stack_get_skill or tech-lead-stack_get_skill) with skillName: "pm-$1" before proceeding.**
+
+1. **Phase 1: Skill Acquisition**: Call the `get_skill` tool:
+   - skillName: "pm-newsletter-generator"
+   - projectName: "<PROJECT_NAME>"
+   - model: "Gemini"
+   - agent: "PM-Assistant"
+
+2. **Phase 2: Newsletter Generation**: Follow the acquired skill's workflow to synthesize git history into a product-focused newsletter.

package/.agents/pm-workflows/pm-progress-translator.md

@@ -0,0 +1,18 @@
+---
+name: pm-progress-translator
+description: Client-Facing Technical Progress Translator
+modes:
+  - write
+---
+
+// turbo
+
+**MANDATORY: You MUST call the get_skill tool (which may be prefixed as mcp_tech-lead-stack_get_skill or tech-lead-stack_get_skill) with skillName: "pm-$1" before proceeding.**
+
+1. **Phase 1: Skill Acquisition**: Call the `get_skill` tool:
+   - skillName: "pm-progress-translator"
+   - projectName: "<PROJECT_NAME>"
+   - model: "Gemini"
+   - agent: "PM-Assistant"
+
+2. **Phase 2: Update Translation**: Follow the acquired skill's workflow to translate technical achievements into user-centric updates for clients or stakeholders.

package/.agents/pm-workflows/pm-release-note-drafter.md

@@ -0,0 +1,18 @@
+---
+name: pm-release-note-drafter
+description: Automated Feature Release Note Drafter
+modes:
+  - write
+---
+
+// turbo
+
+**MANDATORY: You MUST call the get_skill tool (which may be prefixed as mcp_tech-lead-stack_get_skill or tech-lead-stack_get_skill) with skillName: "pm-$1" before proceeding.**
+
+1. **Phase 1: Skill Acquisition**: Call the `get_skill` tool:
+   - skillName: "pm-release-note-drafter"
+   - projectName: "<PROJECT_NAME>"
+   - model: "Gemini"
+   - agent: "PM-Assistant"
+
+2. **Phase 2: Draft Generation**: Follow the acquired skill's workflow to automatically draft release notes from recently merged features.

package/.agents/pm-workflows/pm-risk-detector.md

@@ -0,0 +1,18 @@
+---
+name: pm-risk-detector
+description: Code-Based Milestone Risk Detector
+modes:
+  - write
+---
+
+// turbo
+
+**MANDATORY: You MUST call the get_skill tool (which may be prefixed as mcp_tech-lead-stack_get_skill or tech-lead-stack_get_skill) with skillName: "pm-$1" before proceeding.**
+
+1. **Phase 1: Skill Acquisition**: Call the `get_skill` tool:
+   - skillName: "pm-risk-detector"
+   - projectName: "<PROJECT_NAME>"
+   - model: "Gemini"
+   - agent: "PM-Assistant"
+
+2. **Phase 2: Risk Audit**: Follow the acquired skill's workflow to identify technical risks and bottlenecks threatening the upcoming milestone.

package/.agents/pm-workflows/pm-story-augmenter.md

@@ -0,0 +1,18 @@
+---
+name: pm-story-augmenter
+description: Technical User Story Detail Augmenter
+modes:
+  - write
+---
+
+// turbo
+
+**MANDATORY: You MUST call the get_skill tool (which may be prefixed as mcp_tech-lead-stack_get_skill or tech-lead-stack_get_skill) with skillName: "pm-$1" before proceeding.**
+
+1. **Phase 1: Skill Acquisition**: Call the `get_skill` tool:
+   - skillName: "pm-story-augmenter"
+   - projectName: "<PROJECT_NAME>"
+   - model: "Gemini"
+   - agent: "PM-Assistant"
+
+2. **Phase 2: Story Polish**: Follow the acquired skill's workflow to enhance the user story with technical depth, edge cases, and technical acceptance criteria.

package/.agents/pm-workflows/pm-task-specifier.md

@@ -0,0 +1,18 @@
+---
+name: pm-task-specifier
+description: Auto-Draft Technical Task Specifier
+modes:
+  - write
+---
+
+// turbo
+
+**MANDATORY: You MUST call the get_skill tool (which may be prefixed as mcp_tech-lead-stack_get_skill or tech-lead-stack_get_skill) with skillName: "pm-$1" before proceeding.**
+
+1. **Phase 1: Skill Acquisition**: Call the `get_skill` tool:
+   - skillName: "pm-task-specifier"
+   - projectName: "<PROJECT_NAME>"
+   - model: "Gemini"
+   - agent: "PM-Assistant"
+
+2. **Phase 2: Spec Drafting**: Follow the acquired skill's workflow to generate high-fidelity technical specifications for new features.

package/.agents/workflows/accessibility-audit.md

@@ -0,0 +1,30 @@
+---
+description: Specialized audit for Web Accessibility (A11y).
+modes:
+  - write
+---
+
+// turbo-all
+
+**CRITICAL: PHASE 0 - SKILL ACQUISITION IS NON-NEGOTIABLE.**
+**YOU MUST CALL THE GET_SKILLS TOOL EVEN IF YOU ALREADY HAVE THE CONTEXT. FAILURE TO DO SO BYPASSES MISSION TELEMETRY.**
+
+> [!IMPORTANT]
+> **ANTI-CONFLATION DIRECTIVE:**
+> This file (`.agents/workflows/accessibility-audit.md`) is a workflow launcher stub, NOT the skill definition.
+> Viewing this file via `view_file` does NOT satisfy Phase 0. You MUST call the MCP `get_skills` or `get_skill` tool FIRST with all 4 required fields before executing any other steps.
+
+- **Skill Usage Enforcement (NON-NEGOTIABLE):**
+  - **FORBIDDEN:** Direct file access via `view_file` or `run_command` is strictly prohibited for skill reading.
+  - **IDE / MCP-enabled Agent:** You MUST call the MCP `get_skills` tool (which may be prefixed as `mcp_tech-lead-stack_get_skills` or `tech-lead-stack_get_skills` depending on client prefixing).
+  - **Chat UI (/chat):** You MUST call the internal `get_skill` tool.
+
+1. **Phase 0: Skill Acquisition**: Call the `get_skills` tool (which may be prefixed as `mcp_tech-lead-stack_get_skills` or `tech-lead-stack_get_skills` depending on client prefixing):
+   - skillName: "accessibility-auditor"
+   - projectName: "<YOUR_CURRENT_PROJECT_NAME>"
+   - model: "<YOUR_MODEL_NAME>"
+   - agent: "<YOUR_AGENT_NAME>"
+
+2. **Phase 1: Environment Discovery**: Identify the tech stack by reading root configuration files (e.g., package.json, pyproject.toml, go.mod, Cargo.toml, pom.xml, build.gradle) to understand architectural constraints.
+
+3. Follow its workflow to perform a deep accessibility audit and generate `A11Y_AUDIT.md`.

package/.agents/workflows/ask.md

@@ -0,0 +1,44 @@
+---
+name: ask
+description: A Q&A workflow to chat with the Agent about the codebase.
+modes:
+  - read-only
+---
+
+// turbo
+
+> [!CAUTION] **CRITICAL: MANDATORY READ-ONLY WORKFLOW & SKILL**
+> This workflow and the associated skill are strictly **READ-ONLY**.
+> Under **NO** circumstances may the agent edit, update, delete, or create any code files in the IDE, workspace, or web app.
+> All codebase modifications, file writes, and mutating commands are **STRICTLY PROHIBITED**.
+> The agent is strictly prohibited from using the following tools under this workflow:
+> - `write_to_file`
+> - `replace_file_content`
+> - `multi_replace_file_content`
+> - `run_command` (except for purely read-only commands like `git log`, `cat`, etc., with no side-effects)
+> - `browser_subagent` (except for purely read-only viewing of web app without clicking mutating actions)
+>
+> The agent must act ONLY as a **READ-ONLY ADVISORY ORACLE** and may only output explanation, guidelines, and copy-pasteable code snippets to the chat for manual implementation by the developer.
+
+**CRITICAL: PHASE 0 - SKILL ACQUISITION IS NON-NEGOTIABLE.**
+**YOU MUST CALL THE GET_SKILLS TOOL EVEN IF YOU ALREADY HAVE THE CONTEXT. FAILURE TO DO SO BYPASSES MISSION TELEMETRY.**
+
+> [!IMPORTANT]
+> **ANTI-CONFLATION DIRECTIVE:**
+> This file (`.agents/workflows/ask.md`) is a workflow launcher stub, NOT the skill definition.
+> Viewing this file via `view_file` does NOT satisfy Phase 0. You MUST call the MCP `get_skills` or `get_skill` tool FIRST with all 4 required fields before executing any other steps.
+
+- **Skill Usage Enforcement (NON-NEGOTIABLE):**
+  - **FORBIDDEN:** Direct file access via `view_file` or `run_command` is strictly prohibited for skill reading.
+  - **IDE / MCP-enabled Agent:** You MUST call the MCP `get_skills` tool (which may be prefixed as `mcp_tech-lead-stack_get_skills` or `tech-lead-stack_get_skills` depending on client prefixing).
+  - **Chat UI (/chat):** You MUST call the internal `get_skill` tool.
+
+1. **Phase 0: Skill Acquisition**: Call the `get_skills` tool (which may be prefixed as `mcp_tech-lead-stack_get_skills` or `tech-lead-stack_get_skills` depending on client prefixing):
+   - skillName: "ask"
+   - projectName: "<YOUR_CURRENT_PROJECT_NAME>"
+   - model: "<YOUR_MODEL_NAME>"
+   - agent: "<YOUR_AGENT_NAME>"
+
+2. **Phase 1: Environment Discovery**: Identify the tech stack by reading root configuration files (e.g., package.json, pyproject.toml, go.mod, Cargo.toml, pom.xml, build.gradle) to understand architectural constraints.
+
+3. Follow its workflow to provide architectural insights and manually implementable snippets. **CRITICAL: This workflow is strictly READ-ONLY. The agent is forbidden from updating or altering code in any way.**

package/.agents/workflows/audit-tech-debt.md

@@ -0,0 +1,31 @@
+---
+name: audit-tech-debt
+description: Technical Debt Audit
+modes:
+  - write
+---
+
+// turbo-all
+
+**CRITICAL: PHASE 0 - SKILL ACQUISITION IS NON-NEGOTIABLE.**
+**YOU MUST CALL THE GET_SKILLS TOOL EVEN IF YOU ALREADY HAVE THE CONTEXT. FAILURE TO DO SO BYPASSES MISSION TELEMETRY.**
+
+> [!IMPORTANT]
+> **ANTI-CONFLATION DIRECTIVE:**
+> This file (`.agents/workflows/audit-tech-debt.md`) is a workflow launcher stub, NOT the skill definition.
+> Viewing this file via `view_file` does NOT satisfy Phase 0. You MUST call the MCP `get_skills` or `get_skill` tool FIRST with all 4 required fields before executing any other steps.
+
+- **Skill Usage Enforcement (NON-NEGOTIABLE):**
+  - **FORBIDDEN:** Direct file access via `view_file` or `run_command` is strictly prohibited for skill reading.
+  - **IDE / MCP-enabled Agent:** You MUST call the MCP `get_skills` tool (which may be prefixed as `mcp_tech-lead-stack_get_skills` or `tech-lead-stack_get_skills` depending on client prefixing).
+  - **Chat UI (/chat):** You MUST call the internal `get_skill` tool.
+
+1. **Phase 0: Skill Acquisition**: Call the `get_skills` tool (which may be prefixed as `mcp_tech-lead-stack_get_skills` or `tech-lead-stack_get_skills` depending on client prefixing):
+   - skillName: "technical-debt-auditor"
+   - projectName: "<YOUR_CURRENT_PROJECT_NAME>"
+   - model: "<YOUR_MODEL_NAME>"
+   - agent: "<YOUR_AGENT_NAME>"
+
+2. **Phase 1: Environment Discovery**: Identify the tech stack by reading root configuration files (e.g., package.json, pyproject.toml, go.mod, Cargo.toml, pom.xml, build.gradle) to understand architectural constraints.
+
+3. Follow its workflow to perform a structural health scan and identify ROI-prioritized cleanup tasks.

package/.agents/workflows/changelog.md

@@ -0,0 +1,31 @@
+---
+name: changelog
+description: Generate Changelog
+modes:
+  - write
+---
+
+// turbo-all
+
+**CRITICAL: PHASE 0 - SKILL ACQUISITION IS NON-NEGOTIABLE.**
+**YOU MUST CALL THE GET_SKILLS TOOL EVEN IF YOU ALREADY HAVE THE CONTEXT. FAILURE TO DO SO BYPASSES MISSION TELEMETRY.**
+
+> [!IMPORTANT]
+> **ANTI-CONFLATION DIRECTIVE:**
+> This file (`.agents/workflows/changelog.md`) is a workflow launcher stub, NOT the skill definition.
+> Viewing this file via `view_file` does NOT satisfy Phase 0. You MUST call the MCP `get_skills` or `get_skill` tool FIRST with all 4 required fields before executing any other steps.
+
+- **Skill Usage Enforcement (NON-NEGOTIABLE):**
+  - **FORBIDDEN:** Direct file access via `view_file` or `run_command` is strictly prohibited for skill reading.
+  - **IDE / MCP-enabled Agent:** You MUST call the MCP `get_skills` tool (which may be prefixed as `mcp_tech-lead-stack_get_skills` or `tech-lead-stack_get_skills` depending on client prefixing).
+  - **Chat UI (/chat):** You MUST call the internal `get_skill` tool.
+
+1. **Phase 0: Skill Acquisition**: Call the `get_skills` tool (which may be prefixed as `mcp_tech-lead-stack_get_skills` or `tech-lead-stack_get_skills` depending on client prefixing):
+   - skillName: "changelog-generator"
+   - projectName: "<YOUR_CURRENT_PROJECT_NAME>"
+   - model: "<YOUR_MODEL_NAME>"
+   - agent: "<YOUR_AGENT_NAME>"
+
+2. **Phase 1: Environment Discovery**: Identify the tech stack by reading root configuration files (e.g., package.json, pyproject.toml, go.mod, Cargo.toml, pom.xml, build.gradle) to understand architectural constraints.
+
+3. Follow its workflow to transform git history into semantic release notes.

package/.agents/workflows/clean-code-audit.md

@@ -0,0 +1,31 @@
+---
+name: clean-code-audit
+description: Clean Code Audit
+modes:
+  - write
+---
+
+// turbo
+
+**CRITICAL: PHASE 0 - SKILL ACQUISITION IS NON-NEGOTIABLE.**
+**YOU MUST CALL THE GET_SKILLS TOOL EVEN IF YOU ALREADY HAVE THE CONTEXT. FAILURE TO DO SO BYPASSES MISSION TELEMETRY.**
+
+> [!IMPORTANT]
+> **ANTI-CONFLATION DIRECTIVE:**
+> This file (`.agents/workflows/clean-code-audit.md`) is a workflow launcher stub, NOT the skill definition.
+> Viewing this file via `view_file` does NOT satisfy Phase 0. You MUST call the MCP `get_skills` or `get_skill` tool FIRST with all 4 required fields before executing any other steps.
+
+- **Skill Usage Enforcement (NON-NEGOTIABLE):**
+  - **FORBIDDEN:** Direct file access via `view_file` or `run_command` is strictly prohibited for skill reading.
+  - **IDE / MCP-enabled Agent:** You MUST call the MCP `get_skills` tool (which may be prefixed as `mcp_tech-lead-stack_get_skills` or `tech-lead-stack_get_skills` depending on client prefixing).
+  - **Chat UI (/chat):** You MUST call the internal `get_skill` tool.
+
+1. **Phase 0: Skill Acquisition**: Call the `get_skills` tool (which may be prefixed as `mcp_tech-lead-stack_get_skills` or `tech-lead-stack_get_skills` depending on client prefixing):
+   - skillName: "clean-code"
+   - projectName: "<YOUR_CURRENT_PROJECT_NAME>"
+   - model: "<YOUR_MODEL_NAME>"
+   - agent: "<YOUR_AGENT_NAME>"
+
+2. **Phase 1: Environment Discovery**: Identify the tech stack by reading root configuration files (e.g., package.json, pyproject.toml, go.mod, Cargo.toml, pom.xml, build.gradle) to understand architectural constraints.
+
+3. Follow its workflow to audit architecture and recommend SOLID improvements.

package/.agents/workflows/code-review.md

@@ -0,0 +1,38 @@
+---
+name: code-review
+description: Pre-PR Quality Gatekeeper Code Review
+modes:
+  - write
+---
+
+// turbo-all
+
+**CRITICAL: PHASE 0 - SKILL ACQUISITION IS NON-NEGOTIABLE.**
+**YOU MUST CALL THE GET_SKILLS TOOL EVEN IF YOU ALREADY HAVE THE CONTEXT. FAILURE TO DO SO BYPASSES MISSION TELEMETRY.**
+
+> [!IMPORTANT]
+> **ANTI-CONFLATION DIRECTIVE:**
+> This file (`.agents/workflows/code-review.md`) is a workflow launcher stub, NOT the skill definition.
+> Viewing this file via `view_file` does NOT satisfy Phase 0. You MUST call the MCP `get_skills` or `get_skill` tool FIRST with all 4 required fields before executing any other steps.
+
+- **Skill Usage Enforcement (NON-NEGOTIABLE):**
+  - **FORBIDDEN:** Direct file access via `view_file` or `run_command` is strictly prohibited for skill reading.
+  - **IDE / MCP-enabled Agent:** You MUST call the MCP `get_skills` tool (which may be prefixed as `mcp_tech-lead-stack_get_skills` or `tech-lead-stack_get_skills` depending on client prefixing).
+  - **Chat UI (/chat):** You MUST call the internal `get_skill` tool.
+
+1. **Phase 0: Skill Acquisition**: Call the `get_skills` tool (which may be prefixed as `mcp_tech-lead-stack_get_skills` or `tech-lead-stack_get_skills` depending on client prefixing):
+   - skillName: "code-review-checklist"
+   - projectName: "<YOUR_CURRENT_PROJECT_NAME>"
+   - model: "<YOUR_MODEL_NAME>"
+   - agent: "<YOUR_AGENT_NAME>"
+
+2. **Phase 1: Environment Discovery**: Identify the tech stack by reading root configuration files (e.g., package.json, pyproject.toml, go.mod, Cargo.toml, pom.xml, build.gradle) to understand architectural constraints.
+
+3. Follow its workflow to run a high-density logic and quality audit on the branch.
+
+4. **MANDATORY FOR AUDITORS**: When performing an automated audit, you MUST:
+   - Create a new branch named `audit/<original-branch>`.
+   - Commit your findings and any suggested fixes to this new branch.
+   - Create a Pull Request targeting the original branch.
+   - Assign the person who triggered the audit as the primary reviewer.
+

package/.agents/workflows/competitive-analysis.md

@@ -0,0 +1,46 @@
+---
+name: competitive-analysis
+description: >-
+  Port of the blog's /competitive-analysis - compare this stack against external
+  sources and queue accepted ideas.
+modes:
+  - write
+---
+
+// turbo
+
+**CRITICAL: PHASE 0 - SKILL ACQUISITION IS NON-NEGOTIABLE.**
+**YOU MUST CALL THE GET_SKILLS TOOL EVEN IF YOU ALREADY HAVE THE CONTEXT. FAILURE TO DO SO BYPASSES MISSION TELEMETRY.**
+
+> [!IMPORTANT]
+> **ANTI-CONFLATION DIRECTIVE:**
+> This file (`.agents/workflows/competitive-analysis.md`) is a workflow launcher stub, NOT the skill definition.
+> Viewing this file via `view_file` does NOT satisfy Phase 0. You MUST call the MCP `get_skills` or `get_skill` tool FIRST with all 4 required fields before executing any other steps.
+
+- **Skill Usage Enforcement (NON-NEGOTIABLE):**
+  - **FORBIDDEN:** Direct file access via `view_file` or `run_command` is strictly prohibited for skill reading.
+  - **IDE / MCP-enabled Agent:** You MUST call the MCP `get_skills` tool (which may be prefixed as `mcp_tech-lead-stack_get_skills` or `tech-lead-stack_get_skills` depending on client prefixing).
+  - **Chat UI (/chat):** You MUST call the internal `get_skill` tool.
+
+1. **Phase 0: Skill Acquisition**: Call the `get_skills` tool (which may be prefixed as `mcp_tech-lead-stack_get_skills` or `tech-lead-stack_get_skills` depending on client prefixing):
+   - skillName: "competitive-analysis"
+   - projectName: "<YOUR_CURRENT_PROJECT_NAME>"
+   - model: "<YOUR_MODEL_NAME>"
+
+2. **Phase 1: Source Ingestion**:
+   - Accept URLs, local files, transcripts. Use optional Firecrawl integration for external links.
+   - Summarize each source in <=10 lines (PARAPHRASED), quoting is limited to short attributed fragments.
+
+3. **Phase 2: Practice Extraction**:
+   - Extract specific competitive practices found in the sources.
+   - Format: `practice | paraphrased evidence | source section pointer`.
+
+4. **Phase 3: Four-Pillars Gap Matrix**:
+   - Compare practices to Phase 0 self-inventory.
+   - Format: `practice | pillar(s) | our status (Better / Parity / Gap / N-A) | our artifact path | adoption cost S/M/L | verdict (adopt / decline / investigate)`.
+   - **HARD RULE:** Auto-decline any practice conflicting with any pillar (G-Stack Ethos, MinimumCD, Agent Skills, Modern Web Guidance).
+
+5. **Phase 4: Outputs**:
+   - Write the report to `.dev-team/competitive/YYYY-MM-DD-<slug>.md`.
+   - For each "adopt" verdict, append a drafted `gh issue create --label competitive-analysis` to `.dev-team/inbox.md`.
+   - For each "adopt" verdict, generate a one-line `rtk run reflexion-loop -- "<brief>"` brief.

package/.agents/workflows/design-requirements-to-architecture.md

@@ -0,0 +1,31 @@
+---
+name: design-requirements-to-architecture
+description: Feature Design Assistant
+modes:
+  - write
+---
+
+// turbo-all
+
+**CRITICAL: PHASE 0 - SKILL ACQUISITION IS NON-NEGOTIABLE.**
+**YOU MUST CALL THE GET_SKILLS TOOL EVEN IF YOU ALREADY HAVE THE CONTEXT. FAILURE TO DO SO BYPASSES MISSION TELEMETRY.**
+
+> [!IMPORTANT]
+> **ANTI-CONFLATION DIRECTIVE:**
+> This file (`.agents/workflows/design-requirements-to-architecture.md`) is a workflow launcher stub, NOT the skill definition.
+> Viewing this file via `view_file` does NOT satisfy Phase 0. You MUST call the MCP `get_skills` or `get_skill` tool FIRST with all 4 required fields before executing any other steps.
+
+- **Skill Usage Enforcement (NON-NEGOTIABLE):**
+  - **FORBIDDEN:** Direct file access via `view_file` or `run_command` is strictly prohibited for skill reading.
+  - **IDE / MCP-enabled Agent:** You MUST call the MCP `get_skills` tool (which may be prefixed as `mcp_tech-lead-stack_get_skills` or `tech-lead-stack_get_skills` depending on client prefixing).
+  - **Chat UI (/chat):** You MUST call the internal `get_skill` tool.
+
+1. **Phase 0: Skill Acquisition**: Call the `get_skills` tool (which may be prefixed as `mcp_tech-lead-stack_get_skills` or `tech-lead-stack_get_skills` depending on client prefixing):
+   - skillName: "feature-design-assistant"
+   - projectName: "<YOUR_CURRENT_PROJECT_NAME>"
+   - model: "<YOUR_MODEL_NAME>"
+   - agent: "<YOUR_AGENT_NAME>"
+
+2. **Phase 1: Environment Discovery**: Identify the tech stack by reading root configuration files (e.g., package.json, pyproject.toml, go.mod, Cargo.toml, pom.xml, build.gradle) to understand architectural constraints.
+
+3. Follow its workflow to translate requirements into technical specifications that align with the detected ecosystem.