一键在 Manus 中运行任何 Skill

dashboard

星标2

分支1

更新时间2026年6月16日 10:45

Refresh and open the Agent OS dashboard — a static HTML mission-control page covering today's plan and activity, work across projects, sessions, Kevin's brain, reports, capabilities, and system internals. A refresh snapshots the latest Claude Code sessions (where-am-i radar) and regenerates both dashboard.html and projects/TASKS.md from current state, without the heavy sync work (no compile, flywheel, or briefing). Use when the user says "refresh the dashboard", "update the dashboard", "open the dashboard", or wants the big picture of what Kevin is and what's going on.

安装

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。

在 Manus 中运行

来源

AgentLayer1

AgentLayer1/agent-kevin

打开 GitHub 仓库查看创作者相关仓库

下载

在 Manus 中运行

Dashboard

The Agent OS dashboard is the command center: a self-contained dashboard.html at the agent home, regenerated from current state on demand. No server, zero external requests.

A refresh is intentionally light. Almost everything the dashboard shows — tasks (projects/TASKS.md is rebuilt from frontmatter), skills, settings, persona, reports, git activity, knowledge stats, compile state, logs — is read live from disk every time the dashboard tool runs, so it's always current with no extra step. The one piece of derived state the tool does NOT regenerate is the session radar (the latest Claude Code sessions), so a refresh runs where-am-i first to freshen that, then regenerates. It deliberately skips the heavy sync work (compile, flywheel, briefing, prune, lint-fix) — sync owns those and finishes with these same two steps.

Run

Freshen the session radar — invoke the where-am-i skill (via the Skill tool, default 24h window). It owns the radar end to end: scans the latest sessions, writes the per-session summaries, and persists the report to reports/radar/. One source of truth — don't reimplement its steps here.

Skip this step if a where-am-i radar report was already written in this run or in the last ~15 minutes (e.g. sync just ran, or it's in today's reports list) — don't redo work sync already did; go straight to step 2.
Regenerate the dashboards — one call refreshes both dashboard.html and projects/TASKS.md (the two derived views always regenerate together, and pick up the radar from step 1):

mcp__plugin_agent-kevin_kevin__dashboard

Returns { path, bytes, tasks }.

Open it for the user (macOS) — ONLY if this session's Bash tool runs commands unsandboxed. If the Bash tool description mentions a command sandbox, app launches will fail: skip this step entirely (no open, no open -a, no browser launchers) and just include the file:// path in your reply so the user can open it themselves.

open "<path from step 2>"

If open errors anyway, same rule: do NOT retry with other launchers.

Reply with one line: the dashboard is refreshed (and open, if the launch worked), plus anything the health badge would flag (the tool result alone doesn't carry health — skip unless asked).

Pages

today (plan / goals / today-so-far / news) · tasks (agenda / needs attention) · projects · sessions · brain (threads / memory / concepts / pipeline / lint) · reports · capabilities (cheatsheet / skills / tools / commands / reflexes) · persona · system (context / settings / logs) · profile (reached via the operator card) · status (reached via the health badge). Sub-tabs deep-link: dashboard.html#tasks/attention.

Notes

Near-pure: the dashboard tool never mutates knowledge or task files (projects/TASKS.md is a derived view, rebuilt from frontmatter). The only write a refresh makes is the where-am-i radar report in reports/radar/, written by the where-am-i skill (step 1) — and that step is skipped when a fresh one already exists.
Secrets are redacted at the source (••••); never un-redact them.
Markdown links open via the configured opener app: set the MARKDOWN_URL env var in .claude/settings.local.json ({path} placeholder, e.g. markedit://open?path={path}); defaults to obsidian://open?path={path}&paneType=tab (new Obsidian tab, dashboard stays put).
Every /agent-kevin:sync also refreshes the dashboard; kevin dashboard does the same from a terminal.

同仓库更多 Skills

同仓库

sync

AgentLayer1/agent-kevin

End-to-end refresh — compile pending raw inputs, lint+fix the wiki, run a flywheel pass across active projects, surface what needs attention (including a pending plugin upgrade and any planning/review skill that's come due, with the slash command to run it), optionally chain into a morning or evening briefing, snapshot recent Claude Code sessions (where-am-i radar), then refresh both dashboards (TASKS.md + dashboard.html) last so they capture the briefing's news and the run's final state, and close with a short interview offering concrete next steps (only when something's actually surfaced) that you can act on now or queue as a task. Run anytime you want to bring Kevin's state fully current and get one consolidated update. Heavier than quick-pulse, lighter than running each skill by hand.

2026-07-092

init

AgentLayer1/agent-kevin

Guided first-run onboarding for the agent-kevin plugin. Walks through Kevin's character (SOUL), role (IDENTITY), your basics (name, timezone), an optional web pull from your blog/site/LinkedIn/etc., and communication style — then scaffolds CLAUDE.md (operating manual + @-imports), SOUL.md, IDENTITY.md, USER.md, .claude/settings.json, and seeds four system-architecture concept articles into knowledge/concepts/. If a CLAUDE.md already exists at the home directory, Kevin's version is written to CLAUDE.local.md instead. Skill packs are configured inline at the end or via /agent-kevin:configure-skills any time later. Invoke once after installing the plugin.

2026-07-092

setup-worktree

AgentLayer1/agent-kevin

Create a git worktree for parallel agent work and bootstrap it so it's ready to code — copies the gitignored local files (`.env*`, `.claude/settings.local.json`, `.cursor`, `.cmux`) from the main checkout, installs dependencies, and builds the packages. Use whenever the user asks to spin up a worktree, work on a branch in parallel, set up an isolated checkout for another agent, or "make a worktree for <feature>". First pins down WHICH repo the worktree is for (the user's words, the `$KEVIN_CODE_PATH` default when they assume you know, or by asking when neither resolves), then creates the worktree as a sibling of that repo, never nested inside it, and offers to add it to a sibling `*.code-workspace` if one exists.

2026-07-092

evening-briefing

AgentLayer1/agent-kevin

Tailored evening wrap — what shipped, what drafted, what stalled, goal-delta, and tomorrow's first move. Strict today-only scope; cheeky one-liner when the day is genuinely empty.

2026-07-062

flywheel

AgentLayer1/agent-kevin

Cross-project work session. Triage active tasks, advance every project meaningfully, close what's done, log what mattered. Invoke when you have time to work across the whole portfolio rather than one focus area.

2026-07-062

monthly-goals

AgentLayer1/agent-kevin

Set this month's themes and big-rocks goals — reads the full task board + last 2 weeks of sessions, then interviews you (2-3 rounds) before drafting themes. Run on the 1st of the month or whenever feels right.

2026-07-062

name	dashboard
description	Refresh and open the Agent OS dashboard — a static HTML mission-control page covering today's plan and activity, work across projects, sessions, Kevin's brain, reports, capabilities, and system internals. A refresh snapshots the latest Claude Code sessions (where-am-i radar) and regenerates both dashboard.html and projects/TASKS.md from current state, without the heavy sync work (no compile, flywheel, or briefing). Use when the user says "refresh the dashboard", "update the dashboard", "open the dashboard", or wants the big picture of what Kevin is and what's going on.
allowed-tools	mcp__plugin_agent-kevin_kevin__dashboard, Skill(agent-kevin:where-am-i), Bash

Dashboard

The Agent OS dashboard is the command center: a self-contained dashboard.html at the agent home, regenerated from current state on demand. No server, zero external requests.

Run

Freshen the session radar — invoke the where-am-i skill (via the Skill tool, default 24h window). It owns the radar end to end: scans the latest sessions, writes the per-session summaries, and persists the report to reports/radar/. One source of truth — don't reimplement its steps here.

Skip this step if a where-am-i radar report was already written in this run or in the last ~15 minutes (e.g. sync just ran, or it's in today's reports list) — don't redo work sync already did; go straight to step 2.
Regenerate the dashboards — one call refreshes both dashboard.html and projects/TASKS.md (the two derived views always regenerate together, and pick up the radar from step 1):

mcp__plugin_agent-kevin_kevin__dashboard

Returns { path, bytes, tasks }.

Open it for the user (macOS) — ONLY if this session's Bash tool runs commands unsandboxed. If the Bash tool description mentions a command sandbox, app launches will fail: skip this step entirely (no open, no open -a, no browser launchers) and just include the file:// path in your reply so the user can open it themselves.

open "<path from step 2>"

If open errors anyway, same rule: do NOT retry with other launchers.

Reply with one line: the dashboard is refreshed (and open, if the launch worked), plus anything the health badge would flag (the tool result alone doesn't carry health — skip unless asked).

Notes

Near-pure: the dashboard tool never mutates knowledge or task files (projects/TASKS.md is a derived view, rebuilt from frontmatter). The only write a refresh makes is the where-am-i radar report in reports/radar/, written by the where-am-i skill (step 1) — and that step is skipped when a fresh one already exists.
Secrets are redacted at the source (••••); never un-redact them.
Markdown links open via the configured opener app: set the MARKDOWN_URL env var in .claude/settings.local.json ({path} placeholder, e.g. markedit://open?path={path}); defaults to obsidian://open?path={path}&paneType=tab (new Obsidian tab, dashboard stays put).
Every /agent-kevin:sync also refreshes the dashboard; kevin dashboard does the same from a terminal.