Menu
SkillsMP has collected 15 skills from samchon/AutoMovie. Open a skill to review its source and details.
Showing 15 of 15 collected skills.
Defines automovie's committed contract traceability from product requirements through package-independent system specifications to public TypeScript exports, including @ttsc/evidence populations, citations, exclusions, README participation, stable anchors,…
Defines how automovie runs an ad-hoc experiment: creating a disposable source-linked sandbox under experimental/, briefing and steering a live Claude Code or Codex session that drives its MCP surface against working-tree code, and deciding what an observation…
Defines how packages/cli/scaffold is maintained as the authoring harness every generated project inherits verbatim, including the shipped production skill, the reserved discovery, obligation, and principle inventory, and the negative-probe and…
Defines how this project's production is researched, authored, evidenced, implemented, and reviewed, including the mutually exclusive film, brief, and library shapes, the settings-to-screenplay ladder, the model, space, material, instance, motion, and system…
Defines automovie implementation rules, testing standards, the always-100% coverage mandate, validation, consequence analysis, and change integrity. Use before writing or modifying source, tests, workflows, package wiring, or fixtures.
Defines the design rules for packages/mcp: what the tool surface is and is not allowed to do for its client, server/tool arrangement as an ongoing experiment, and the hard JSDoc-length constraints MCP clients impose. Use before adding or reshaping an MCP tool…
Defines the automovie product contract, what the product deliberately does not do, the long-haul mission, workspace layout, and canonical commands. Use when orienting in the repository, working inside any package, choosing a build, test, format, or coverage…
Defines how to drive the viewer/playground through the Playwright library to inspect renders, poses, and motion against expectation, including how to reach a real GPU context and how to compare against previous behavior without disturbing a shared checkout.…
Defines the .wiki/ working knowledge base, package README, code JSDoc, and agent-instruction conventions for automovie. Use before writing or modifying docs, AGENTS.md, or a SKILL.md, and revise the wiki as the work proceeds, not at the end.
Defines the repository-wide issue campaign for automovie: exhaustive solo discovery, wiki-backed candidate adjudication, main-agent-vetted issue publication, then DAG-ordered parallel implementation by one owner per issue in one checkout and one branch, each…
Defines exhaustive review, Self-Review, and solo repository-wide issue-discovery rounds for automovie. Use for every self-review or unqualified review request and as the review mode inside issue campaigns, where each parallel issue owner reviews its own…
Defines what automovie models and what it refuses to model, and the verification discipline every geometry, parameter, and derived-data change is held to. Use before any model, geometry, rig, morph, or asset-pipeline work, and before proposing anything that…
Defines the isolated-topology variants of automovie review and issue campaigns: parallel discovery, parallel review, and implementation partitioned into separate worktrees, branches, and pull requests. Use only when the user explicitly requests a team,…
Defines automovie branch, commit, pull-request, check, and merge workflows. Use when shipping a topic-unit PR under the standing instruction, when the user asks to open, update, or merge a pull request, or when a standing autonomous mandate authorizes…
Runs structured multi-agent discussions for open-ended automovie topics. Use only when the user explicitly asks for Discussion or a team debate with transcripts and conclusions; do not use for Self-Review, Review Cycle, research review, or issue discovery.