菜单
Prepare a repository with the directories and gitignore entries that Accelerator skills expect. Safe to run repeatedly.
用 Codex 或 Claude 帮你安装 复制这段 Prompt，粘贴到 Codex、Claude 或其他助手里，让它检查 Skill 页面并帮你完成安装。
基于 SOC 职业分类
Open the accelerator meta visualiser. Launches the companion-window server in the background and returns a URL. Subcommands stop and status manage the running server.
Apply pending Accelerator meta-directory migrations to bring a repo into line with the latest plugin schema. Destructive by default but guarded — refuses to run on a dirty working tree and prints a one-line preview per pending migration before applying.
Generate a structured design inventory for a frontend source — tokens, components, screens, and features — by crawling it with code analysis, live Playwright inspection, or both. Use when you need to capture a snapshot of a current or target design surface before running analyse-design-gaps. Produces a dated artifact directory with an inventory.md and screenshots/. Re-running for the same source-id supersedes the prior snapshot without losing it.
Compare two design inventories produced by inventory-design and emit a structured gap artefact whose prose paragraphs satisfy the extract-work-items cue-phrase contract. Use after running inventory-design for both a current and target design surface. The resulting gap artefact under meta/design-gaps/ feeds directly into /accelerator:extract-work-items to produce actionable work items.
Conduct comprehensive codebase research by spawning parallel subagents and synthesising findings into a research document. Use when the user needs to deeply understand a codebase area or answer technical questions.
View, create, or edit Accelerator plugin configuration. Manage document templates.
| name | init |
| description | Prepare a repository with the directories and gitignore entries that Accelerator skills expect. Safe to run repeatedly. |
| argument-hint | (no arguments — safe to run repeatedly) |
| disable-model-invocation | true |
| allowed-tools | ["Bash(${CLAUDE_PLUGIN_ROOT}/scripts/config-*)"] |
You prepare a consumer repository with all directories and gitignore entries that Accelerator skills expect. This is safe to run repeatedly — it reports what was created versus what already existed.
Resolve each output directory using the plugin's path configuration:
Plans directory: !${CLAUDE_PLUGIN_ROOT}/scripts/config-read-path.sh plans
Research directory: !${CLAUDE_PLUGIN_ROOT}/scripts/config-read-path.sh research_codebase
Decisions directory: !${CLAUDE_PLUGIN_ROOT}/scripts/config-read-path.sh decisions
PRs directory: !${CLAUDE_PLUGIN_ROOT}/scripts/config-read-path.sh prs
Validations directory: !${CLAUDE_PLUGIN_ROOT}/scripts/config-read-path.sh validations
Review plans directory: !${CLAUDE_PLUGIN_ROOT}/scripts/config-read-path.sh review_plans
Review PRs directory: !${CLAUDE_PLUGIN_ROOT}/scripts/config-read-path.sh review_prs
Review work items directory: !${CLAUDE_PLUGIN_ROOT}/scripts/config-read-path.sh review_work
Work items directory: !${CLAUDE_PLUGIN_ROOT}/scripts/config-read-path.sh work
Notes directory: !${CLAUDE_PLUGIN_ROOT}/scripts/config-read-path.sh notes
Design inventories directory: !${CLAUDE_PLUGIN_ROOT}/scripts/config-read-path.sh research_design_inventories
Design gaps directory: !${CLAUDE_PLUGIN_ROOT}/scripts/config-read-path.sh research_design_gaps
Global directory: !${CLAUDE_PLUGIN_ROOT}/scripts/config-read-path.sh global
Execute the following steps in order, tracking what was created versus what already existed so you can present a summary at the end.
bash "${CLAUDE_PLUGIN_ROOT}/skills/config/init/scripts/init.sh"
The script creates 13 meta/ directories with .gitkeep files, creates the
.accelerator/ core scaffold (.gitignore, state/, tmp/, skills/,
lenses/, templates/), writes the inner tmp .gitignore, and appends the
.accelerator/config.local.md rule to the root .gitignore. It is idempotent
— safe to run repeatedly.
Only after all steps above have completed successfully, present a summary:
Initialisation complete:
Directories:
✓ {plans directory} (created | already exists)
✓ {research directory} (created | already exists)
✓ {decisions directory} (created | already exists)
✓ {prs directory} (created | already exists)
✓ {validations directory} (created | already exists)
✓ {review plans directory} (created | already exists)
✓ {review prs directory} (created | already exists)
✓ {review work items directory} (created | already exists)
✓ {work items directory} (created | already exists)
✓ {notes directory} (created | already exists)
✓ {design inventories directory} (created | already exists)
✓ {design gaps directory} (created | already exists)
✓ {global directory} (created | already exists)
Accelerator scaffold:
✓ .accelerator/.gitignore (created | already exists)
✓ .accelerator/state/.gitkeep (created | already exists)
✓ .accelerator/skills/.gitkeep (created | already exists)
✓ .accelerator/lenses/.gitkeep (created | already exists)
✓ .accelerator/templates/.gitkeep (created | already exists)
✓ .accelerator/tmp/.gitignore (created | already exists)
✓ .accelerator/tmp/.gitkeep (created | already exists)
Gitignore entries:
✓ .accelerator/config.local.md (added | already present)
Use the actual resolved paths in the output (not the variable names).
/accelerator:init bootstraps a fresh repository. To upgrade an existing repository after a plugin update, use /accelerator:migrate instead — it applies ordered, idempotent migrations to bring meta/, .claude/, and .accelerator/ in line with the latest schema.