init
Initialize KERNEL links and AgentDB in a confirmed Vaults directory.
- License 许可
- Apache-2.0
- 最近更新
- 2026-11-11
- Stars
- ★ 1.6k
- Forks
- 474
01 前言 Front Matter
- name
- init
- kernel
- { "kind": "operator", "version": 1, "confirmation": "always", "side_effects": "writes_repo" }
- description
- Initialize KERNEL links and AgentDB in a confirmed Vaults directory.
- allowed-tools
- Bash
- user-invocable
- true
- disable-model-invocation
- true
02 资源文件 1
- 参考文档agents/openai.yaml203 B
03 SKILL.md 正文
KERNEL init
Explicit, once-per-machine setup. This skill never moves or replaces the whole
~/.claude directory.
Requirements
Git, SQLite 3, jq, Python 3, and Bash. Install the plugin first with the exact ID:
/plugin marketplace add ariaxhan/kernel-claude
/plugin install kernel@kernel-marketplace
/reload-plugins
Detection and confirmation
Locate the runtime selector at
$HOME/.claude/plugins/cache/kernel-marketplace/kernel/current, source its
hooks/scripts/common.sh, and call kernel_validate_runtime_root on the selected
root. Stop if validation fails.
Call detect_vaults. Its order is: an existing valid KERNEL_VAULTS,
$HOME/Documents/Vaults, $HOME/Vaults, $HOME/Downloads/Vaults, then the
reported $HOME/Documents/Vaults fallback.
Before writing, show the exact Vaults path and ask:
Set up KERNEL in this Vaults directory?
Continue only after confirmation. A different path must be exported as
KERNEL_VAULTS and detected again.
Writes
After confirmation, create only these directories when absent:
$VAULTS/_meta/agentdb
$VAULTS/_meta/research
$VAULTS/_meta/plans
$VAULTS/_meta/handoffs
$VAULTS/_meta/checkpoints
$VAULTS/_meta/retrospectives
$VAULTS/_meta/agents
$VAULTS/_meta/logs
$VAULTS/.claude/kernel
$VAULTS/.local/bin
Call kernel_update_current, then use kernel_init_host_link for exactly:
$VAULTS/.local/bin/agentdb -> $CACHE/current/orchestration/agentdb/agentdb
$VAULTS/.claude/kernel/orchestration -> $CACHE/current/orchestration
$VAULTS/.claude/kernel/hooks -> $CACHE/current/hooks
Missing links may be created. Correct links are unchanged. Recognizable numbered KERNEL links may be repaired. A regular file, directory, malformed link, or unrelated link is preserved and stops setup with an actionable message. Never force-replace links.
Initialize the database through the shared helper, which pins AgentDB to the selected Vaults even when init was invoked from another directory:
kernel_init_agentdb "$VAULTS" "$CACHE"
Do not edit a shell startup file automatically. If $VAULTS/.local/bin is not on
PATH, print this line for the user to add to their chosen shell configuration:
export PATH="${KERNEL_VAULTS:-$HOME/Documents/Vaults}/.local/bin:$PATH"
Verify
"$VAULTS/.local/bin/agentdb" status
readlink "$HOME/.claude/plugins/cache/kernel-marketplace/kernel/current"
readlink "$VAULTS/.local/bin/agentdb"
Report every directory and link created. Existing AgentDB data, manifests, receipts, project configuration, and unrelated files are not rewritten.
Legacy ~/.claude consolidation
This is not part of init. Moving an existing ~/.claude can affect plugins,
credentials, settings, and other tools. If a user explicitly requests consolidation,
make a dated backup, show a complete move plan and rollback command, and ask for a
separate confirmation before any move. There is no one-line migration.
Recovery
- Missing plugin: install
kernel@kernel-marketplace, then reload. - Wrong Vaults: export
KERNEL_VAULTSto the existing root and rerun init. - Unsafe host path: inspect the named object; KERNEL will not overwrite it.
- Intentional runtime rollback: run
scripts/select-runtime.sh /path/to/runtimefrom a trusted KERNEL checkout. It selects code only; it does not convert state formats.
04 同仓库更多 Skills
e2e
Writes, runs, and debugs end-to-end tests for web and mobile apps with e2e, the agentic test runner, which mixes agent.act goals with exact locator assertions on a Playwright browser or an iOS or Android device. Also drives a project's running app live over the e2e MCP server to look at a screen, check a UI change, take a screenshot or record a video, and try a locator; and runs bug bashes, parallel e2e explore runs with each bug proven by a repro test. Use when a project depends on e2e or has an e2e.config.ts, when asked for end-to-end, browser, mobile, or agentic UI tests, to check a change in the running app, to QA or bug bash an app or branch, or when an e2e run fails.
create-verification-skill
Generate a project-local verify-<app> skill on top of e2e, so any coding agent can launch the app, check it is worth driving, exercise a feature the way a user does, keep evidence, and bug bash it. Writes the skill, a feature map with one explore charter per feature, and proves it once. Use for /create-verification-skill, "make a verification skill for this repo", or when a project has no scripted way to prove UI behavior.
deja-search
deja-vu memory — search the user's past AI coding sessions with the deja CLI. Holds what they already settled and never wrote into the repo (naming rules, banned or already-tried tools, required checks before merge, preferred values, who owns what). Use before your first edit in a task, before you call a change done or ready to merge, before you recommend a tool, library, design or reviewer, before re-debugging an error that may already be solved, when they say 'didn't we fix this before', 'what did we decide about X' or 'I already have that', and before telling them something on this machine does not exist.
deja-reflect
End-of-session reflection and memory curation with deja-vu. Review what was decided, what commands broke and what repaired them, and record durable project rules or decisions with deja remember so the next session starts with them.
revid
Make, edit, export and post AI videos with Revid (revid.ai) through the Revid MCP server. Use whenever the user wants a short-form or long-form video made (from an idea, a script, a URL, an article, a PDF, a tweet, a product page, their own clips, or a song), wants to find what is going viral on TikTok, YouTube, Instagram or X before making something, wants to edit or re-export an existing Revid project, generate a standalone AI image or clip, make a series episode, or publish and schedule a finished video to connected social accounts. Covers setup (hosted MCP with OAuth, or an API key), the 36 tools, credits, polling and the rules for publishing. Routes to the more specific revid-* skills when one fits.
review-study
Review a QuestionPunk survey before launch for wording, answer choices, branching, and missing settings. Use for check my survey or review this draft before publishing.