capturing-ui-evidence
The mechanical capture contract for UI verification, driving Playwright CLI against the pinned dev server URL to collect a full-page screenshot, computed styles and an accessibility snapshot.
creating-implementation-plans
Writes TSH's implementation plan as a committed file with phased tasks that each name their files and a verifiable Definition of Done, plus a verification document for the final phase.
debugging-code
Finds and fixes the root cause of a bug by reproducing it, securing a failing test, naming the causal chain with evidence and measuring the reach of the fix before applying it minimally.
discovering-technical-context
Establishes which conventions a change must follow before it is written, in priority order from the plan's persisted context to project instructions, codebase patterns and external documentation.
orchestrating-feature-implementation
Implements a feature, ticket or plan end to end from the main conversation by delegating each plan task to implementer subagents and closing with parallel code review and functional verification.
reviewing-code
TSH's evidence-based pre-merge review of a delivered change set, comparing it with the plan, running the project's checks and applying anti-pattern, security and scalability checks.
reviewing-ui
Runs one UI verification pass against Figma by delegating fresh capture to ui-capture-worker and the design comparison to ui-reviewer, returning PASS, FAIL or VERIFICATION NOT RUN.
verifying-ui
The judging standard for comparing implemented UI against its Figma design across five categories under strict tolerances, ending in PASS, FAIL or VERIFICATION NOT RUN.