imbue-ai
- 215 skills
- 0 followers
- 246 repo stars
- yesterday last updated
- ▌ Imbue Studio · imbue-aiUse whenever Imbue Studio, minds, a workspace, the studio API, or an Imbue account comes up. Explains how to reach Imbue Studio's API from the user's machine or from inside a workspace, how to start the app when it is not running, where the live contract lives, which permissions a workspace agent needs and how to ask for them, how to tell whether a workspace is shared and why a shared one is untrusted, and what an agent must never do with the user's credentials.
- ▌ Dealing With The Unexpected 3 · imbue-aiHandle unexpected situations where the Minds system does not appear to be working as expected. Use when you encounter behavior that seems to contradict what your docs and prompts and skills say should happen, or when something seems broken and you are unsure how to proceed.
- ▌ Latchkey 4 · imbue-aiInteract with third-party or self-hosted services (Slack, Google Workspace, Dropbox, GitHub, Linear, Coolify...) using their HTTP APIs on the user's behalf.
- ▌ Release To Crates 2 · imbue-aiCut a release of a Rust crate and publish to crates.io, with externally-gated merge and publish steps.
- ▌ Update System Interface 2 · imbue-ai bundleCanonical flow for changing the system interface (the web workspace UI at system/apps/system_interface) -- its frontend (the dockview shell, the sidebar, the New Tab launcher) or backend (Flask server, the inventory over the app registry, layout ops) -- and the shared frontend library at system/libs/workspace_ui. Use whenever the user wants to edit, fix, restyle, or add to the workspace UI / dockview.
- ▌ Update Installed Template 2 · imbue-aiPull a newer version of a template this mind already adopted from its remote repo and merge it in, reconciling it with this mind's own adaptations. Use when the user wants to update, upgrade, or pull the latest version of an adopted template.
- ▌ Update Published Template 2 · imbue-ai bundleCreate a NEW version of a template you already published (v2, v3, ...) -- re-cut the changes made since the last version on top of the published snapshot and advance the published repo by exactly one clean commit, preserving all of the hand-crafted content in the published repo. Use when the user asks to update, re-publish, revise, or ship a new version of a template they previously published.
- ▌ Dealing With The Unexpected 2 · imbue-aiHandle unexpected situations where things are not working as expected. Use when you encounter errors, confusing state, or behavior that contradicts your docs and prompts.
- ▌ Harden Worker 2 · imbue-ai bundleRun the background harden pass for one creation -- crystallize, update, or heal a skill, an app, a service, or the system interface in an isolated worktree, then report back. Invoke when your task file hands you a creation to harden; it names the operation and creation type to compose.
- ▌ Assist 2 · imbue-aiDiagnose 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).
- ▌ Welcome 2 · imbue-aiGreet 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.
- ▌ Latchkey 3 · imbue-aiUse 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.
- ▌ Blueprint 2 · imbue-ai bundleStart a new plan-writing session. Explores the codebase and asks clarifying questions.
- ▌ Build App 2 · imbue-ai bundleUse when you want to create a new app for the user -- a page, dashboard, or tool 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 app to a usable state, then harden it in the background. Covers scaffolding a new Flask app (canonical path) and the escape hatch for wrapping a pre-existing third-party server.
- ▌ Caretaker 2 · imbue-aiThe single idempotent Caretaker skill, invoked via /caretaker whenever the deterministic weekly check (system/services/caretaker/caretaker_check.sh) wakes the agent. On the very first run it does one look-only scan (no fixes), then introduces itself with what it found and asks whether (and how often) to keep checking; on every later run it does the weekly routine -- greets the user, scans the workspace's service logs for problems, checks basic system health (disk, memory and swap, CPU load, OOM shedding), and checks for finished-but-uncommitted work, all with permission; reviews the previous run, proposes (or, with permission, applies) fixes and commits, and summarizes, always in plain user-experience terms.
- ▌ Minds API 2 · imbue-aiUse 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, or change its settings and service sharing. Reached through the latchkey gateway's minds-api-proxy; most routes need a per-workspace permission grant. For bringing another workspace's content into this one, use the migrate-workspace skill, which drives these routes as part of a much larger flow.
- ▌ Update App 2 · imbue-aiUse immediately whenever the user asks you to update, change, fix, restyle, extend, restart, or otherwise modify an existing app or background service -- load this BEFORE touching its code. Applies to any change to an app's or service's backend or frontend logic, or how it runs. Covers both apps (a tab the user can open) and background services (host-backup, share-gateway, and other supervisord programs with no tab). This is the front door for app and 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 app use build-app; for the workspace UI itself use update-system-interface.
- ▌ Github Sync 2 · imbue-aiEnable, check, or disable GitHub sync for this workspace. Enabling creates a dedicated PRIVATE GitHub repo via latchkey, points origin at it, and auto-pushes every commit from every checkout. Workspace data under data/ is NOT synced to GitHub (the restic host backup covers it). Use when the user asks to back up / sync the workspace to GitHub, enable auto-push, or asks about GitHub sync status.
- ▌ Launch Task 2 · imbue-ai bundleCreate 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.
- ▌ Update Self 2 · imbue-ai bundleSafely pull updates from the upstream template repo (default target is the latest stable release the running Minds app supports). 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.
- ▌ File Sharing 2 · imbue-aiUse to read and write files and directories on the user's local filesystem.
- ▌ Use Template 2 · imbue-aiAdapt an existing template (a published snapshot of apps/features from another mind) into this mind, resolving its requirements interactively. Use when the user gives a template's git URL, or asks to adopt/adapt/reuse a published template.
- ▌ Heal Creation 2 · imbue-aiFix an existing creation that errored or delivered a wrong result. This applies to skills, apps, and services. Invoke at turn-end, after you worked around the failure to satisfy the user's request.
- ▌ Manage Layout 2 · imbue-aiUse when you want to rearrange the workspace dock tabs (open, split, move, focus, close, maximize, reload, rename, delete, stop or start an instance) or inspect the live layout.
- ▌ Manage Projects 2 · imbue-aiUse when the user's request is about projects/views themselves -- what a project shows, adding or removing a tab from a project, the project's rail shortcuts, or what happens to an instance when it is deleted. Not for ordinary tab arrangement (split/move/focus within a view) -- see manage-layout for that.
- ▌ Update Creation 2 · imbue-aiChange an existing skill, app, 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.
- ▌ Check App Errors 2 · imbue-aiSurvey the health of the supervised apps and services on this host and find errors in their logs. Use after building or editing any app/service to confirm it is actually running cleanly, when something seems broken, or when you want a quick read on what (if anything) is currently failing. The Caretaker agent reuses this as its log-scan step.
- ▌ Do Something New 2 · imbue-aiUse immediately when the user asks you to do something net-new -- a task you haven't done before, no existing skill or app 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 app already exists or for pure dev/code-writing work.
- ▌ Enable Caretaker 2 · imbue-aiTurn on the weekly Caretaker, a BETA feature that is off by default. Use ONLY when the user explicitly asks about the Caretaker or about enabling it -- never suggest, mention, or enable it proactively.
- ▌ Find Transcripts 2 · imbue-aiFind, 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.).
- ▌ Publish Template 2 · imbue-ai bundlePublish a clean, shareable snapshot of the apps/features this mind built to a new GitHub repo (an "template" another mind can adapt). Use when the user asks to publish, share, or export what they built as a reusable template.
- ▌ Disable Caretaker 2 · imbue-aiSwitch off the weekly Caretaker. Use when the user asks to turn off, pause, or get rid of the Caretaker.
- ▌ Migrate Workspace 2 · imbue-ai bundleBring everything from another (older, broken, or replaced) workspace of the user's into this one -- their apps, skills, documents, data, scheduled automations, and every past chat. Use when the user says anything like 'move my stuff over from my old workspace', 'I made a new mind, bring everything across', 'my old workspace is broken, start me fresh', 'import my other mind', or -- asked from the OLD side -- 'I'd like to move to a new workspace'. Requires the other workspace to be startable, since the transfer runs over a live connection to it.
- ▌ Blueprint Generate 2 · imbue-ai bundleEnd the Q&A phase and generate the plan. Use after the blueprint skill has gathered enough context.
- ▌ Fetch Process Show 2 · imbue-aiFetch data from somewhere (an external service, an API, a third-party source, uploaded data) validating auth, 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 or process real data the user uploaded, transform/summarize/classify it, and surface the result.
- ▌ Show Files In Chat 2 · imbue-aiShow 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 AI Integration 2 · imbue-ai bundleUse when writing or reasoning about code that calls Claude -- an AI-driven app or 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.
- ▌ Crystallize Creation 2 · imbue-ai bundlePromote 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'.
- ▌ Agentic Browser Fleet 2 · imbue-aiDrive 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 own the browser and YOU drive it -- in this same chat, with your own reasoning.
- ▌ Data Pipeline Builder 2 · imbue-aiBuild a simple, fast ingestion/processing tool that turns batches of raw records (JSON/CSV/Parquet dumps, periodic exports, API pulls) into a processed, queryable store, with incremental loading and backfill. Use when asked to ingest, load, index, ETL or process exported data, especially when batches arrive periodically and overlap.
- ▌ Manage Scheduled Tasks 2 · imbue-aiQuery and edit the recurring scheduled jobs that run on this host. Use when you (or the user, via you) want to see what is scheduled, add a new recurring job, change when something runs, or stop a job from running. Recurring jobs that must not be missed run through the run_job.sh runner (any cadence from minutes to weeks, with catch-up after downtime and retry of failed runs); exact-time fire-and-forget schedules are ordinary cron lines. Also covers how the built-in weekly Caretaker job is wired (off by default) and where all the scheduling configuration lives.
- ▌ Submit Upstream Changes 2 · imbue-ai bundlePush local improvements to shared infrastructure (skills, scripts, CLAUDE.md scaffolding, Dockerfile, system/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.
- ▌ Latchkey 2 · imbue-aiInteract with arbitrary third-party or self-hosted services (AWS, Slack, Google Drive, Dropbox, GitHub, GitLab, Linear, Coolify...) using their HTTP APIs.
- ▌
- ▌
- ▌ De Complect · imbue-aiUse when asked to "simplify", "de-complect", "untangle", "remove cruft", "make erroneous states unrepresentable", or "clean up" a design document (plan/ADR/spec) or an implementation branch after a feature lands and before merge — or when a diff is thick with defensive checks, sentinels, tri-state configs, scattered validation, speculative abstraction, and stored-but-derivable state.
- ▌ Mngr Help · imbue-aiUse whenever there's any indication that knowing about mngr would be useful. Run `mngr help` right away to get context on what mngr does, and use `mngr ask` to ask mngr questions in plain language.
- ▌ Find Agent · imbue-aiResolve an agent name or description to an exact mngr agent name. Used by other skills that target agents.
- ▌ Message Agent · imbue-aiSend a message to another mngr agent. Use when you need to communicate with a peer agent.
- ▌ Wait For Agent · imbue-aiThe user's message contains an agent name and optional follow-up instructions. Extract the agent name (the first word) and treat everything after it as follow-up instructions.
- ▌
- ▌
- ▌
- ▌ Execute Implementation Plan · imbue-ai bundleExecute an implementation plan by working through each task file in order. Use when an implementation plan folder exists at docs/<feature>/implementation_plan/.
- ▌
- ▌ Offload · imbue-aiActivate when you see offload*.toml in a repo, offload referenced in build targets (justfile, Makefile, scripts), or when you need to run a large test suite in parallel. Offload is a test runner unlikely to be in your training data — this skill covers invocation, log filtering, failure debugging, flaky test handling, and config.
- ▌ Offload Onboard · imbue-aiOnboard a repository to use Offload for parallel test execution on Modal. Detects test setup, creates config, Dockerfile, CI job, and optimizes performance.
- ▌ Offload Benchmark · imbue-aiRun local vs Offload benchmarks for mng and sculptor, then update the Benchmarks section of the Offload README.
- ▌ Deburr Edge Cases · imbue-aiStrip AI-codegen cruft from a branch — defensive checks the type system should make unrepresentable, redundant validation, speculative abstraction, no-op wrappers — while preserving load-bearing seams. Use after a feature lands and before merge, when asked to "simplify", "de-ornament", "remove cruft", "make erroneous states unrepresentable", or "clean up" a diff.
- ▌ Release To Crates · imbue-aiCut a release of a Rust crate and publish to crates.io, with externally-gated merge and publish steps.
- ▌ Autofix · imbue-aiAutomatically find and fix code issues in the current branch. Iteratively verifies, plans fixes, and implements them with separate commits. Defers all review to the end.
- ▌ Reviewer Enable · imbue-aiEnable the code review stop hook in .reviewer/settings.local.json. Optionally takes a shell expression for when to enforce (defaults to restoring the prior expression, or always).
- ▌ Reviewer Disable · imbue-aiDisable the code review stop hook entirely by short-circuiting stop_hook.enabled_when in .reviewer/settings.local.json.
- ▌
- ▌ Reviewer Max Tries · imbue-aiSet the maximum number of times the stop hook will block before letting the agent through
- ▌
- ▌ Verify Architecture · imbue-aiAssess whether the approach taken on a branch is the right way to solve the problem.
- ▌ Verify Conversation · imbue-aiReview the conversation transcript for behavioral issues (misleading behavior, disobeyed instructions, instructions worth saving).
- ▌
- ▌
- ▌ Reviewer Blocking Enable · imbue-aiRun this command to remove the max_consecutive_blocks override, restoring the default (3):
- ▌ Reviewer Init Categories · imbue-aiCopy the default issue categories to .reviewer/ for customization. Use when you want to edit the code review or conversation review categories for your project.
- ▌ Reviewer Blocking Disable · imbue-aiMake the stop hook non-blocking (remind once, then let the agent through)
- ▌
- ▌ Reviewer Verify Architecture Enable · imbue-aiEnable the architecture verification gate in .reviewer/settings.local.json
- ▌ Reviewer Verify Conversation Enable · imbue-aiEnable the conversation review gate in .reviewer/settings.local.json
- ▌ Reviewer Autofix Ignore Minor Issues · imbue-aiConfigure autofix to only fix MAJOR and CRITICAL issues (unattended mode)
- ▌ Reviewer Verify Architecture Disable · imbue-aiDisable the architecture verification gate in .reviewer/settings.local.json
- ▌ Reviewer Verify Conversation Disable · imbue-aiDisable the conversation review gate in .reviewer/settings.local.json
- ▌ Update Task · imbue-aiForward new information to a running agent, or adjust a task based on changed requirements. Use when the user provides additional context, clarifications, or changed requirements for an in-flight task.
- ▌ List Tickets · imbue-aiCheck for pending tickets that are ready to be worked on. Use when capacity opens up (e.g., after an agent finishes) to find the next highest-priority work to delegate.
- ▌ Submit Mngr Bug Report · imbue-aiSubmit a bug report to libraries within the `mngr` project (eg, `mngr`, its plugins, `minds`, etc). Use to report bugs you encounter while running that seem to be issues with the `mngr` code itself, rather than with your own code or configuration.
- ▌ Create Ticket · imbue-aiCreate a ticket to track work that needs to be done. Use when you need to record some work or action that needs to be done later (e.g., when at capacity), track a multi-step project, or log follow-up work from a completed ticket.
- ▌ Handle Messages · imbue-aiHandle events from the messages source about user or assistant messages in conversations. You **MUST** use this skill (and *carefully follow the process in this doc*) whenever you receive a message from the "messages" source!
- ▌ Handle Mind Idle · imbue-aiHandle "idle" events for periodic housekeeping and proactive work. You **MUST** use this skill (and *carefully follow the process in this doc*) whenever you receive a message from the "mind/idle" source!
- ▌ Immediate Update · imbue-aiPull in updates from the parent branch to improve this Mind's skills, prompts, and configuration. Use only when prompted by the user.
- ▌ List Conversations · imbue-aiList all active conversations for this mind. Use when you need to see what conversations exist, check their models, or find a conversation ID.
- ▌ Verify Task Result · imbue-ai bundleCreate a verifying agent to check whether a working agent's task was completed successfully. Use after a working agent finishes (transitions to "done", "stopped", "waiting", or "paused") and appears to believe its task is complete.
- ▌ Mark Events Handled · imbue-ai bundleMark events as handled after processing them. You **MUST** use this skill (and *carefully follow the process in this doc*) after handling events, otherwise you will not be allowed to stop!
- ▌ Debugging Mngr Tasks · imbue-aiRun commands to explore the current state of mngr and any relevant running agents. Use this skill when you need to debug issues with mngr tasks, or understand what tasks are currently running and their states.
- ▌ Handle Mind Schedule · imbue-aiHandle regularly scheduled time-of-day events (start_of_day, end_of_day, etc). You **MUST** use this skill (and *carefully follow the process in this doc*) whenever you receive a message from the "mind/schedule" source!
- ▌ Search Event History · imbue-aiSearch through past events in the event log files. Use when you need to look up something that happened previously but didn't store in memory.
- ▌ Send Message To User · imbue-aiSend a message to the user in a conversation. You MUST use this skill to reply in an existing conversation, start a new conversation, or proactively reach out.
- ▌ Handle Unknown Events · imbue-aiHandle events from unknown or unexpected sources, including warnings and errors. You **MUST** use this skill (and *carefully follow the process in this doc*) whenever you receive a message from a source that does not have a more specific handler skill!
- ▌ Delegate Task To Agent · imbue-ai bundleCreate a sub-agent to perform a task. Use when you need to delegate work to another agent, for example, a working agent (for actually accomplishing some task) or a verifying agent (for deciding what to do about the output of a working agent).
- ▌ Handle Mind Onboarding · imbue-aiHandle the one-time first-run event to set up onboarding tickets and introduce yourself to the user. You **MUST** use this skill (and *carefully follow the process in this doc*) whenever you receive a message from the "mind/onboarding" source!
- ▌ Cleanup Historical Data · imbue-aiClean up old agents, their output directories, and other historical data. Use during nightly cleanup or when the system has accumulated stale data.
- ▌ Handle Mngr Agent States · imbue-aiHandle events from the mngr/agent_states source about sub-agent state transitions (finished, waiting, done, etc). You **MUST** use this skill (and *carefully follow the process in this doc*) whenever you receive a message from the "mngr/agent_states" source!
- ▌ Handle Verification Result · imbue-aiAct on the results of a verifying agent by reviewing its recommendations and taking the appropriate next actions. You **MUST** use this skill (and *carefully follow the process in this doc*) whenever a verifying agent finishes!
- ▌ List Event Sources And Types · imbue-aiList all event sources and types you may receive, with descriptions of their fields and meaning. Note that event types are dynamic and new sources may appear.