Stian Haga
Community@Shtian · Oslo
Stian Haga's skill registry provides software engineering playbooks, design principles, and verification methods for code review, debugging, migrations, and technical documentation.
Agent Skills by Stian Haga
Showing 34 vetted skills indexed across 1 GitHub repositories.
Shtianfigure-it-out
Designs an auditable multi-phase workflow for large migrations and ambiguous engineering tasks.
Shtianplaywright-cli
Automates browser interactions for web testing, form filling, screenshots, and data extraction.
Shtianinterrogate
Spawns multiple LLM reviewers to adversarially review code changes and synthesizes a verdict.
Shtianprinciple-boundary-discipline
Concentrates validation and error handling at system boundaries while keeping business logic in pure functions.
Shtianprinciple-model-the-domain
Guides encoding domain logic into data structures instead of scattered conditionals.
Shtiantechnical-writing
Reviews and rewrites technical documentation using Diátaxis, Google style, STE, and Global English rules.
Shtianprinciple-fix-root-causes
Guides debugging by tracing symptoms to root causes instead of applying symptom fixes.
Shtianwhy
Investigate design rationale behind code by querying git history and MCP evidence sources in parallel.
Shtianprinciple-encode-lessons-in-structure
Converts recurring corrections into lint rules, metadata flags, runtime checks, or scripts.
Shtianhow
Generates architectural explanations of codebases by orchestrating parallel explorer and explainer subagents.
Shtianblast-radius
Analyze what a code change could break beyond the diff and prove safety by running real code.
Shtianprinciple-subtract-before-you-add
Guides removal of dead code and redundant logic before implementing additions or refactors.
Shtianprinciple-foundational-thinking
Guides structural decisions on data structures, scaffolding order, and concurrency before writing logic.
Shtianprinciple-build-the-lever
Guides building rerunnable codemods, scripts, and generators instead of performing edits by hand.
Shtianpoteto-mode
Enforces a disciplined agent workflow with playbooks, principles, and verified subagent delegation.
Shtianprinciple-make-operations-idempotent
Designs state-mutating operations to converge to the same end state across retries and crashes.
Shtianprinciple-sequence-verifiable-units
Sequences multi-step code changes into small units verified before advancing.
Shtianprinciple-test-behavior-not-implementation
Rewrites or deletes tests that assert implementation details instead of observable behavior.
Shtianprinciple-outcome-oriented-execution
Guides planned rewrites and migrations toward target architecture with explicit verification boundaries.
Shtianarchitect
Designs software architecture sketches with types, signatures, and module boundaries before implementation.
Shtianprinciple-redesign-from-first-principles
Redesigns existing code as if new requirements were foundational assumptions from the start.
Shtianprinciple-minimize-reader-load
Reviews code for excessive indirection layers and hidden mutable state to reduce reader cognitive load.
Shtianprinciple-exhaust-the-design-space
Builds competing prototypes to compare alternatives before committing to novel design decisions.
Shtianprinciple-migrate-callers-then-delete-legacy-apis
Migrate internal callers to new APIs and delete legacy paths in one refactor wave.
Frequently Asked Questions About Stian Haga
FAQPage SchemaWhat tasks can I accomplish using Shtian's skills?▼
You can design auditable migration playbooks, run adversarial multi-model code reviews, measure blast radius of diffs, debug to root causes, architect types and module structure, write Diátaxis-standard documentation, and drive browser testing, form filling, screenshots, and web data extraction via playwright-cli.
Who are Shtian's skills designed for?▼
Software engineers, architects, and technical writers handling non-trivial changes: large migrations, stateful domain logic, concurrent shared-state design, code review, postmortems, and documentation. The principle-* skills target anyone wiring validation, designing types, or refactoring in statically-typed languages.
How do the verification and review skills work in practice?▼
Skills like blast-radius and create-verification-skill run real code against the actual artifact rather than relying on self-reports. Interrogate spawns multiple independent reviewers, arena runs parallel candidate solutions and grafts the strongest parts, and figure-it-out logs decisions through a hypothesis loop.
What prerequisites or dependencies do these skills require?▼
Most skills are invocation-disabled playbooks requiring no runtime dependencies. Playwright-cli requires Bash access to the playwright-cli command. The why skill discovers available MCPs across source control, issue trackers, chat, observability, error tracking, and analytics warehouses to gather cited evidence.
Do Shtian's skills cover documentation and writing standards?▼
Yes. The technical-writing skill enforces Diátaxis structure, Google developer style, STE instruction rules, and Global English syntax for docs, RFCs, readmes, PR descriptions, and commit messages. The unslop skill removes AI tells from any writing and must always apply.