default-workspace-template

Run the background harden pass for one artifact -- crystallize, update, or heal a skill, a service, or the system interface in an isolated worktree, then report back. Invoke when your task file hands you an artifact to harden; it names the operation and artifact to compose.

Promote the just-finished work into a new, reusable, committed, tested skill. Use when that process would recur with new inputs -- especially something you figured out through research or debugging -- or when the user says 'crystallize this'.

Fetch data from somewhere (an external service, an API, a third-party source), process it, and show it to the user -- the "go get all this stuff, do something to it, and put it in front of me" task. Use when the ask is to retrieve real data, transform/summarize/classify it, and surface the result. Validates auth first, confirms a real sample covering every data shape, then crystallizes the pipeline in the background while building surfaces.

Use when writing or reasoning about code that calls Claude -- an AI-driven service, an AI integration, or a skill's scripted model step. Covers the three scenarios (one-shot completion, one-shot agentic task, full agent) and the cost / credentialing model.

Fix an existing artifact that errored or delivered a wrong result. This applies to skills or web services. Invoke at turn-end, after you worked around the failure to satisfy the user's request.

Create a sub-agent to perform a larger task. Use when work is large enough to warrant a separate context, involves multi-file changes, or benefits from isolation.

Change an existing skill, service, or shared script/reference (under .agents/shared/) -- extend it, refactor it, or just verify it still works. Invoke at turn-end when a skill ran but you had to do extra repeatable work by hand, or when you and the user discussed a change to it and applied it live.

Publish a clean, shareable snapshot of the apps/features this mind built to a new GitHub repo (an "inspiration" another mind can adapt). Use when the user asks to publish, share, or export what they built as a reusable template.

Use whenever you want to use latchkey commands or interact with third-party or self-hosted services (Slack, Google Workspace, Dropbox, GitHub, Linear, Coolify...) using their HTTP APIs on the user's behalf.

Safely pull updates from the upstream template repo (default target is the latest stable release). Use when you want to incorporate upstream skills, script fixes, or config improvements. For pushing local improvements back upstream, use the `submit-upstream-changes` skill instead.

Handle unexpected situations where things are not working as expected. Use when you encounter errors, confusing state, or behavior that contradicts your docs and prompts.

Canonical flow for changing the system interface (the web workspace UI at apps/system_interface) -- its frontend (dockview shell, chat rendering, progress view) or backend (Flask server, agent discovery, layout ops). Use whenever the user wants to edit, fix, restyle, or add to the workspace UI / chat interface / dockview.

Use when you want to create a new web view for the user -- a page, dashboard, or app they can open as a tab. Runs an interactive flow: confirm the look and feel on a cheap throwaway mock first, then build the real site to a usable state, then harden it in the background. Covers scaffolding a new Flask service (canonical path) and the escape hatch for wrapping a pre-existing third-party server.

Use to act on OTHER Minds workspaces on the user's behalf -- list them, create a fresh one, SSH into one, read or export its backups, start/stop/destroy/recover it. The headline workflow is migrating content out of an old/broken workspace into a new one and then handing the cleanup to the new workspace's own agent. Reached through the latchkey gateway's minds-api-proxy; most routes need a per-workspace permission grant.

End-to-end dev workflow for the minds app stack -- first-time bring-up, every-startup vendor/mngr sync, and the iteration loop against a running Docker agent. Use this when starting or restarting the dev Electron app, or after changing any minds component (mngr, the system interface, the default workspace template).

Adapt an existing inspiration (a published snapshot of apps/features from another mind) into this mind, filling in its holes interactively. Use when the user gives an inspiration's git URL, or asks to adopt/adapt/reuse a published inspiration.

Use immediately whenever the user asks you to update, change, fix, restyle, extend, restart, or otherwise modify an existing service -- load this BEFORE touching the service's code. Applies to any change to a service's backend or frontend logic, or how it runs. Covers both user-facing web services (a tab the user can open) and background daemons (runtime-backup, cloudflared, and other supervisord programs with no tab). This is the front door for service edits: it owns the live change loop (apply the change so it takes effect, refresh the user's view, verify) and hands the change to the turn-end hardening flow. For creating a brand-new web view use build-web-service; for the workspace UI itself use update-system-interface.

Enable, check, or disable GitHub sync for this workspace. Enabling creates a dedicated PRIVATE GitHub repo via latchkey, points origin at it, auto-pushes every commit from every checkout, and continuously syncs runtime/ state (memory, tickets, transcripts). Use when the user asks to back up / sync the workspace to GitHub, enable auto-push, restore a previous workspace's state, or asks about GitHub sync status.

Use to read and write files and directories on the user's local filesystem.

Show a file to the user in chat. Display an image inline (chart, plot, screenshot, diagram, rendered figure, photo), or offer any other file (PDF, CSV, log, zip, spreadsheet, ...) as a download link. Use whenever you have a file on disk you want the user to see or download, or want to embed an image from a public URL.

Use when you want to rearrange the workspace dockview tabs (split, move, focus, rename, close, maximize, reload, swap a URL) or inspect the live layout.

Push local improvements to shared infrastructure (skills, scripts, CLAUDE.md scaffolding, Dockerfile, supervisord.conf) back to the parent template repo so other agents derived from the template benefit. Opens a separate per-feature PR per logical fix; never pushes directly to upstream `main`. Do not push agent-specific content (PURPOSE.md, memory, runtime state). For pulling updates from upstream, use the `update-self` skill instead.

Find, read, or search through any chat message, transcript, or conversation content from this host -- whether from an active agent, a past session, a deleted agent, a sub-agent, or a worker. Use this skill any time a user asks about chat histories or you otherwise want to access them. NOTE: this skill only covers Minds agents -- not other services (ChatGPT, claude.ai, etc.).

Drive a fleet of shared Chromium browsers yourself, one command at a time, from your shell. Use when the user wants you to do something on the web (log in somewhere, fill a form, click through a flow, read a page that needs interaction) rather than just fetch a URL. YOU run the `agentic-browser-fleet` commands, look at the page, decide what to click, and click it -- in this same chat, with your own reasoning.

Diagnose and fix a problem the user is hitting in this workspace, and escalate built-in (non-user) issues to imbue. Invoked as `/assist <description>` by the minds "get help -> have an agent help" flow (also usable directly when the user describes something broken).

Greet the user with a short, friendly welcome message when a new project/agent is first started. Invoked automatically as the first message from the minds desktop client.

Use immediately when the user asks you to do something net-new -- a task you haven't done before, no existing skill or service applies, and getting it right will require nontrivial research, exploration, or experimentation. Routes the request to the right interactive flow. When doing this, give a very short confirmation message to the user's request, then load this immediately before responding further. Your confirmation message shouldn't mention loading the skill. Skip when an applicable skill or service already exists or for pure dev/code-writing work.