baodq97

Decompose a domain into bounded contexts, aggregates, entities, value objects, and domain events with ubiquitous-language naming, mapping bounded contexts to candidate microservice boundaries. Use whenever the user wants to decompose a domain, split a system into bounded contexts, identify aggregates / value objects / domain events, define clean domain entities, draw clear boundaries, or plan a monolith-to-microservices breakup. Also use when an already-decomposed domain has evolved and the user wants to update, extend, re-run, or "read existing and update" an existing `docs/domain` model rather than start over — re-running delta-merges into the existing docs (stable ids, preserves human edits, flags drops) instead of overwriting. Works from a prose domain/requirements description, not existing code. Trigger even when the user never says "DDD", as long as they are carving a domain into boundaries, modelling entities/aggregates/events, or updating a model they already have.

Structure a clear, verifiable /goal from rough user input through guided interview. Use this skill whenever the user says "define goal", "tạo goal", "/goal", "goal cho task này", "I want to do X" (where X is complex/multi-step), "structure this task", "make this into a goal", or when you detect the user has a vague multi-step objective that would benefit from structured goal definition. Also trigger when the user pastes a wall of requirements and needs them organized into a runnable goal. Covers all task types: implementation, exploration/research, audit/review, and refactoring.

Install a skill from a GitHub URL into the user's personal skills directory (~/.claude/skills/). Use whenever the user says "chom <url>", "install this skill from github", "add/import/grab this skill", or pastes a GitHub URL that clearly points at a SKILL.md folder — handle both full-repo skills and subfolder skills like https://github.com/github/awesome-copilot/tree/main/skills/azure-devops-cli. Also handles optional description optimization of the newly installed skill when the user explicitly asks for it afterward. Trigger even when the user doesn't say the word "chom" as long as the intent is "pull a skill from GitHub into my skills folder".

This skill should be used when the user asks to "analyze business requirements", "do business analysis", "create a requirements document", "map business processes", "write user stories", "feasibility assessment", "stakeholder analysis", "data modelling", "process improvement", "acceptance criteria", "business case", "gap analysis", "as-is to-be analysis", "business modelling", "requirements elicitation", "impact analysis", or mentions business analyst / business analysis work. Guides developers, analysts and aspiring BAs through business analysis tasks using SFIA 9 framework with a 5-phase workflow (Discover → Define → Analyze → Design → Deliver & Validate), inline coaching, and competency assessment. Do NOT use for: solution architecture, infrastructure design, product roadmap management, or code-level implementation decisions.

This skill should be used when the user asks to "define IT strategy", "create IT governance", "write an IT roadmap", "do IT leadership planning", "IT governance framework", "review my IT strategy", "CIO task", "IT portfolio", "IT operating model", "IT investment", "digital transformation", "technology strategy", "IT leadership", "service management", "IT budget", "vendor management", "IT risk", "information management", "IT roadmap", or mentions IT leadership or CIO work. Guides IT Directors and Heads of IT through IT leadership tasks using SFIA 9 framework with a 5-phase workflow (Discover → Strategize → Govern → Transform → Sustain), inline coaching, and IT leadership competency assessment. Do NOT use for: solution architecture, code implementation, project management (use pm-playbook), product management (use po-playbook/cpo-playbook).

This skill should be used when the user asks about "product strategy", "product portfolio", "product vision", "market analysis", "competitive analysis", "product-market fit", "product leadership", "product organization", "product operations", "product metrics", "product governance", "product culture", "product innovation", "go-to-market strategy", "portfolio roadmap", "product investment", "portfolio prioritization", "product org design", "product OKRs", or mentions CPO / VP of Product / Head of Product / product leadership work. Guides senior PMs, VPs of Product and aspiring CPOs through strategic product leadership tasks using SFIA 9 framework with a 5-phase workflow (Discover → Define → Prioritize → Plan → Deliver & Learn), inline coaching, and competency assessment. Do NOT use for: day-to-day PO backlog management (use po-playbook), solution architecture (use sa-playbook), IT strategy (use cio-playbook).

This skill should be used when the user asks to "define a technology vision", "create a technology strategy", "write a tech roadmap", "do technology evaluation", "build vs buy analysis", "engineering culture", "platform strategy", "tech stack assessment", "engineering excellence", "technical debt strategy", "R&D strategy", "innovation pipeline", "technology due diligence", or mentions technology leadership work. Guides engineering leads and VPs through CTO tasks using SFIA 9 framework with a 5-phase workflow (Discover → Envision → Build → Scale → Evolve), inline coaching, and technology leadership assessment. Do NOT use for: IT operations/service management (use cio-playbook), solution architecture (use sa-playbook), product management (use cpo-playbook).

This skill should be used when the user asks to "create an enterprise architecture", "define a business capability model", "write architecture principles", "do enterprise architecture", "build a technology roadmap", "architecture governance", "architecture review board", "IT strategy", "target architecture", "reference architecture", "TOGAF", "architecture standards", "capability model", "business architecture", "business capability", or mentions enterprise architecture work. Guides solution architects and tech leads through EA tasks using SFIA 9 framework with a 5-phase workflow (Discover \u2192 Define \u2192 Design \u2192 Govern \u2192 Evolve), inline coaching, and architecture assessment. Do NOT use for: solution-level design (use sa-playbook), code implementation, CI/CD pipeline setup, or project management.

Use the tmem CLI to inspect and manage the local memory store — record/vector counts, keyword/hybrid search, view persona, list scenes, read one full scene block (tmem scene <name>), recent changes, sync vectors, configure thresholds. Also covers CROSS-PROJECT exploration — search every project store at once (tmem search <q> --all), list all memory stores (tmem projects), and collapse legacy cwd-keyed fragment stores into their project root (tmem migrate-fragments). Trigger when the user asks "how many memories do I have", "show my persona", "what scenes exist", "search memories for X", "search across all projects", "what projects/stores do I have", "my memory is fragmented", "open that scene", "recent memory changes", or when you need to check memory state before/after an operation. Do NOT use for extracting or consolidating memories — those have dedicated skills.

Consolidate L1 memory atoms into L2 scene blocks and L3 persona. Triggers when the user says "consolidate memories", "build persona", "update persona", "update scenes", "organize memories", or after memory-seed completes. Also triggers via asyncRewake pipeline after N conversation turns. This skill is about ORGANIZING existing memories into higher structures — for creating memories from transcripts use memory-seed instead.

Extract L1 memory atoms from Claude Code conversation history. Triggers when the user says "seed memories", "extract memories", "backfill memory", "remember my history", "learn from past conversations", "what do you know about me", or after /memory-init when conversation history exists. Also use when asyncRewake pipeline flags pending sessions. This skill is about CREATING new memories from transcripts — for inspecting existing memories use tmem-cli instead.

One-shot entry point for Contributor Intelligence — the user just drops a GitHub link or names an engineer and you take it from there. Use this whenever someone pastes a GitHub profile/repo URL (github.com/<user> or github.com/<user>/<repo>) or a handle and asks to "profile", "analyze how X works", "learn from X", "study this engineer", "phân tích người này", "học từ người này", or asks "how do I use /contrib". Drives the whole pipeline end-to-end (add → ingest → build → playbook), or guides the user through it if they'd rather drive. This is the orchestrator — the per-phase skills (contrib-ingest, contrib-consolidate, contrib-synthesize) do the actual work; do NOT use this one to inspect or synthesize an already-built persona (use contrib-synthesize) or for the self-memory feature.

Consolidate a contributor's L1 atoms into an L3 persona across 11 dimensions. Triggers when the user says "build contributor persona", "consolidate <user>", or after contrib-ingest completes. This ORGANIZES existing atoms — for creating atoms use contrib-ingest.

Extract L1 contributor atoms from a GitHub subject's raw activity. Triggers after "tmem contrib raw <id>" or when the user says "ingest contributor", "profile <user>", "analyze how <user> works". This CREATES atoms from raw GitHub events — for synthesis use contrib-synthesize.

Use this skill when someone wants to LEARN from engineers they've already profiled — extracting actionable lessons from contributor personas. Trigger on intents like: "what do the top SWEs / engineers / contributors I've profiled have in common", "extract their shared / common capabilities", "turn a contributor's profile into copyable habits, heuristics, or a learnable playbook", "what habits should I copy from <person>", or "compare me / my style to <role-model> dimension-by-dimension and find my gap". Also covers surfacing exemplar teaching quotes across profiled engineers. Operates on existing personas (built via contrib-consolidate); cross-engineer comparison or common-capability synthesis needs ≥2 subjects. Do NOT use for: building, ingesting, or inspecting a single persona's raw contents; commit/trajectory stats; summarizing documents; recalling what you know about the user; or playbooks unrelated to engineer profiles (e.g. team incident-response runbooks).

Close a mandat implementation slice into one owner-decision packet before any governed-doc status flip. Use this whenever a code slice has landed on main and one or more governed docs (US/RFC/ADR/PRD) are candidates to advance status (open->in-progress, ->done, proposed->accepted), or when the user asks to verify + reconcile + red-team before ratifying, or says "close this slice", "prep the flips", "run the close workflow", "ready to flip". It fans out a gate-verify, a doc-keeper drift reconcile, and one red-team per flip candidate — so the owner ratifies ONCE from a single packet instead of per-doc. Reach for it any time you are about to propose a mandat status flip; skipping it risks flipping a doc to done whose ACs the code no longer matches.

Use BEFORE opening or proposing any PR — a pre-PR self-review of the diff for reuse, simplification, comment policy, and scope, so review rounds catch design issues instead of mechanical noise. Tier 1, dependency-free.