atelier

Plugin atelier from marketplace skill in IamK77/Skill. Commands read from the repo's manifest — 1 month ago.

Skills (36)

• canon

  Generate visual points of view and unexpected design languages from a product\u2019s tensions and meaning. Use for creative direction, not an upstream certificate required by other lenses.

• color

  Explore palette ideas through perception, material associations, temperature, and semantic relationships. Use for generative colour thinking; contrast measurements are optional instruments, not taste gates.

• form

  Explore shape, depth, light, texture, and material metaphors for an interface. Use to generate visual form, not to enforce universal radius or shadow formulas.

• graphics

  Invent imagery, iconography, illustration, and generative visual languages. Use for expressive possibilities before choosing DOM, SVG, Canvas, or GPU implementation.

• layout

  Invent spatial compositions through attention, rhythm, and alternative information metaphors. Use for layout exploration, not a fixed grid-and-spacing inspection routine.

• motion

  Explore motion as rhythm, continuity, character, and narrative. Use to invent interaction possibilities, not to require animation or enforce one universal motion budget.

• pilot

  Route atelier work to an execution SOP, a creative heuristic, or a direct answer. Use when the needed kind of help is unclear; do not impose a suite-wide pipeline.

• systems

  Turn a selected visual direction into versioned tokens, reusable components, and an inspectable handoff. Use for design-system implementation, not to choose the creative direction by checklist.

• type

  Explore typographic voices, rhythm, and multilingual expression. Use to generate type directions, not to force a modular scale or impose one cultural model of text.

• holdfast

  Warm, playful exploration of shared worlds, distributed systems, and questions that want a stranger frame. Follow an intriguing association beyond the usual protocol menu; make room for unfinished ideas without rushing to appraisal or an implementation plan. Not a distributed-correctness audit.

• aegis

  Run a scoped defensive security review and remediation workflow. Produce a threat model, reproducible findings, tested changes, and residual-risk handoff.

• assay

  Build regression tests around observable behaviour and demonstrate that they detect the fault. Use for test design, missing coverage, and validating a code change.

• flightline

  Prepare a reviewable, reproducible change and delivery pipeline. Use for repository hygiene, CI work, or release preparation without implying permission to commit, push, or deploy.

• gauge

  Install or tune useful static feedback for a codebase and verify that it catches representative mistakes. Use for type-checking and lint feedback loops, not claims that static checks prove correctness.

• groundwork

  Turn a request into an agreed, testable task definition. Use when requirements, constraints, examples, or acceptance boundaries are unclear.

• gungnir

  Investigate a specifically authorised security hypothesis with minimal, reproducible evidence. Use only on an agreed target and scope; deliver a bounded finding rather than speculative accusations.

• husbandry

  Perform evidence-led maintenance and refactoring while preserving required behaviour. Use for cleanup, migration, and reducing an identified maintenance burden.

• load-bearing

  Explore alternative software architectures through boundaries, reversibility, and borrowed models. Use when the design space is still open, not for a mandatory architecture audit.

• plumb

  Generate alternative representations, APIs, and algorithms by changing the way a problem is described. Use for design exploration at code scale, not a compulsory code-quality checklist.

• stationkeeping

  Prepare and execute an explicitly authorised operational change with observability, rollback, and post-change checks. Use for release or incident procedures, not unattended production action.

• crucible

  Invent and reshape research mechanisms through analogy, inversion, and recombination. Use while the idea is unsettled; no mandatory tournament, novelty certificate, or convergence.

• envoy

  Turn a validated research record into a clear manuscript or response. Use for research communication, preserving claim-to-evidence links and explicit author decisions.

• forge

  Execute a defined research experiment with traceable inputs, configurations, runs, and outputs. Use for experiment implementation and reproducibility, not to invent the research claim.

• ledger

  Write a reproducible experimental plan for a chosen research question. Use after direction selection to define comparisons, data boundaries, metrics, and resource limits.

• prospect

  Find research possibilities by connecting tensions, absences, and neighbouring fields. Use for open-ended question generation rather than literature scoring or forced project selection.

• reckoning

  Analyse research results against the recorded plan, inspect failure patterns, and bound the claims. Use for evidence review before writing or making a research decision.

• forage

  Find and capture a bounded shortlist of software projects for a concrete need. Use for repository discovery, with dated sources and an explicit search stopping point.

• touchstone

  Evaluate a specific software project for a stated use using source, maintenance, compatibility, and trial evidence. This is a repository-assessment skill, not the root A/B experiment platform.

• bearings

  Explore frontend architecture by reframing journeys, ownership, and the experience of time. Use before committing a product direction, not as a compulsory frontend architecture checklist.

• bulwark

  Maintain frontend module boundaries and reusable design artifacts with demonstrable checks. Use for architecture erosion, stale abstractions, and design-system integration.

• keel

  Build a walking skeleton for one selected frontend journey across real integration seams. Use to establish an end-to-end path, not for ordinary CSS questions or speculative architecture.

• livery

  Explore coherent visual directions across colour, type, space, form, imagery, and motion. Use to invent a visual character, not to certify polish or run every atelier lens.

• lookout

  Establish delivery observation and decision-linked frontend telemetry for a chosen release. Use for preview, measurement, and post-release checks with privacy and rollback boundaries.

• seaworthy

  Harden a selected frontend journey across async states, recovery, accessibility, and performance constraints. Use for implementation readiness, not a universal visual-style audit.

• trials

  Build frontend regression tests that observe real user behaviour and survive implementation refactors. Use for interaction, integration, and critical journey coverage.

• wellspring

  Generate alternatives for frontend state and interaction models using ownership, time, and representation shifts. Use when the model is awkward, not for a mandatory state audit.