Hanh Ngô
Community@hanh-nd · Hanoi
Hanh Ngô maintains a 25-skill registry covering the full software delivery lifecycle: research, planning, code review, refactoring, testing, and persistent project memory.
Agent Skills by Hanh Ngô
Showing 24 vetted skills indexed across 1 GitHub repositories.
hanh-ndwiki
Compile handoffs and conversations into a persistent project knowledge wiki.
hanh-ndrecall
Search persistent memory for context relevant to the current question.
hanh-ndinit-digest
Initializes optional local conversation digesting for Memory Kit with a specified model.
hanh-ndmake-skill
Creates and validates AI agent skills using test-driven development methodology.
hanh-ndresearch
Conducts multi-source technical research and produces a verified report with confidence-tiered findings.
hanh-ndvalidate
Validates skill outputs against frozen expectations in a bounded PASS/FAILED retry loop.
hanh-ndbrainstorm
Transforms raw ideas into consensus-backed Design Briefs through structured discovery and challenge.
hanh-ndplan
Generates implementation plans, work breakdowns, and test mappings as saved handoff files.
hanh-ndinvestigate
Diagnose bugs and failing tests by verifying root causes with evidence before any fix.
hanh-ndclarify
Resolves ambiguities in acceptance criteria before implementation planning begins.
hanh-ndcode-refactor
Analyzes code structure and produces a refactor proposal without modifying files.
hanh-ndcode-review
Reviews diffs, PRs, and commits with evidence-backed findings across critical and informational categories.
hanh-ndcode
Implements WBS plans and investigation reports by editing files, running tests, and logging decisions.
hanh-nddelegate
Delegates tasks to external agent CLIs like Gemini, Claude, or Codex with optional handoff context.
hanh-ndcode-simplify
Improves code readability through renames, guard clauses, and dead-code removal without changing behavior.
hanh-nddebate
Stress-test prior skill outputs through multi-round adversarial agent debate.
hanh-nde2e-review
Reviews Playwright and Cypress end-to-end test diffs for requirement proof, selector stability, and CI reliability.
hanh-ndreview
Reviews local uncommitted git changes with Jira ticket context and routes diffs to specialized review skills.
hanh-ndreview-pr
Fetches a Bitbucket pull request, checks out the branch, and produces a full code review report.
hanh-ndticket
Fetches a Jira ticket and routes it to the planning pipeline.
hanh-ndpreview
Convert markdown documents into self-contained glanceable HTML pages with diagrams and drill-downs.
hanh-ndlk:init
Initializes a research foundation by aligning user intent and mapping knowledge topics.
hanh-ndlk:roadmap
Generates time-bound learning roadmaps from adversarial multi-agent research of NotebookLM sources.
hanh-ndlk:list
Lists all learning topics with status and timestamps from state JSON files.
Frequently Asked Questions About Hanh Ngô
FAQPage SchemaWhat tasks can I accomplish with Hanh Ngô's skill registry?▼
You can run the full delivery cycle: brainstorm raw ideas, clarify ambiguous acceptance criteria, produce implementation plans and WBS artifacts, investigate bugs, implement code, refactor, simplify, review PRs and E2E test diffs, validate outputs in PASS/FAILED loops, and maintain a persistent project wiki.
Who is the target audience for these skills?▼
Software engineers, tech leads, and reviewers who need structured planning before coding, rigorous semantic code review of features and diffs, E2E test review for Playwright and Cypress suites, and Jira-integrated ticket-to-plan routing within their development process.
How do the review and validation skills work in practice?▼
Append 'with /validate' to any command to run a skill and check its output in a PASS/FAILED loop. Use review-pr to fetch and check out a PR locally, review for uncommitted changes with Jira context, and e2e-review for browser test diffs with selector-stability and CI-reliability judgment.
What prerequisites or external dependencies do these skills require?▼
The delegate skill requires an external agent CLI such as Gemini, Claude, or Codex. The brainstorm and plan skills specify the gemini-3-pro-preview model. Ticket and review skills expect Jira access, and init-digest optionally sets up a local conversation digest for the Memory Kit.
Do these skills modify my codebase directly?▼
Only the code skill edits files in place while implementing a WBS or Investigation Report. code-refactor and code-simplify are analysis-oriented: code-refactor produces a Refactor Proposal without modifying files, and code-simplify improves readability without changing behavior or signatures.