Skills & the Autonomous Skill System

A skill is a package that teaches your AI employee how to perform a specific task. It bundles task instructions, executable code, reference documentation, and assets in one folder so they can be reused.

Skills come in two kinds. Official skills, supported by UpServe, are available to everyone. Skills you create or import from GitHub or a zip are yours alone. There’s no way to publish a skill to other users or share it by link. If you belong to an organization, though, you can share a skill you made as an organization skill so the organization’s assigned AI can use it.

Since May 2026, you rarely need to think about skills at all. Your employee autonomously finds the right skill for the job, surfaces a 1-click install card in chat when something better is available, self-repairs defects in skills it has customized, and periodically proposes library clean-ups when things get cluttered. You step in only at the moments that matter.

How your employee uses a skill

When your employee puts a skill to work, three things happen in order:

Search official skills: As the employee starts a task, it automatically scans the official skills for ones that could help.
1-click install: When a good match is found, an install card appears in chat. One click adds the skill to this employee.
Run immediately: The installed skill is available right away for the next task, with no separate command needed from you.

Note: in scheduled runs, installing new skills from official skills is not allowed. Only skills already added to the employee can be called directly.

At a glance

Question	Answer
Do I have to install skills myself?	No. Your employee finds what it needs and shows a 1-click install card in chat. One button and you’re done.
Who decides which skill to use?	The employee analyses its own task notes, repeated patterns, and failing skill signals, then auto-suggests up to 3 candidates.
What if a skill breaks?	If the employee has its own copy of a skill, it fixes it directly; for an official skill the report goes to the UpServe team, and for your own skill it becomes an improvement proposal for you.
What happens when my library gets messy?	The system periodically checks and bundles any clean-up candidates into a notification (at most once a week per user). Nothing is ever automatically deleted or changed.
What about external services like Notion or GitHub?	Connect once and the skill will receive your connection details securely on every run.
Will skills affect my computer?	No. Skills run on an isolated AI work computer in the cloud; nothing touches your local machine.
Is reading a PDF different from creating one?	Yes. The skill that reads PDFs and pulls out tables comes attached by default, but creating a PDF (or merging, splitting, watermarking, filling in a form, and similar) is a separate official skill. Your employee finds and offers to install it when it’s first needed, or you can just ask in chat, for example “create a PDF for this”.

Seven moments you’ll meet skills

1. A skill suggestion card in chat

When the employee believes a skill would help with what it’s doing, it shows a card inline in the conversation. The card explains why and offers an Install button. One click installs and activates it immediately.

Suggestions are based on three types of signals:

Notes left after completing recent tasks (what worked, what fell short)
Repeated-pattern detection (when the employee keeps reaching for the same tool)
Failing skill signals (when a skill has errored repeatedly)

These signals are combined into a recommendation score, and only skills scoring 75 or above (out of 100) appear in the card. If you never see a card, your employee is handling things well enough on its own.

The “install button never expires” behavior below is not yet live as of this document’s last update. Until it ships, a skill-install approval card times out the same as any other approval (after its 1-hour response window), so you’d need to ask the employee to install again.

The install button on this card is planned to never expire. Once that ships, if the card sits unanswered for a while and the conversation moves on, you’ll still be able to come back to that card later and tap the install button to go through with it. Other kinds of approvals (running code, clicking through a page) time out because the situation may have already changed by the time you respond; installing a skill isn’t like that, since the outcome is the same whenever you tap it. If an install is blocked (for example, the publisher took the skill down or it’s no longer supported), the card shows why.

2. Bringing in a skill from a public GitHub repository

Beyond what your employee suggests, you can also pull a skill straight from a public GitHub repository.

From the Skills screen: In the Skills tab, tap Import from GitHub at the top and enter the repository (owner/repo or a full URL). It’s imported right away. The result is registered as a skill only you can use, which you can then install on any employee you choose.
In chat: You can also just ask your employee with a link, for example “install the skill from this GitHub repo”. In this case the employee first shows you what it found in the repository, and the install only goes through once you approve it. As with an official skill install, this card’s install button is planned to stay usable without expiring.

The repository needs a skill description file , and there are limits on file count and total size. If a repository contains more than one skill, you choose which one to bring in.

3. Connecting an external service once

Skills that work with Notion, GitHub, Slack, or similar services need a one-time login (connect with account). You can start that from the Connections section in your employee’s settings, or directly from a skill card in Official skills. Once connected, the skill receives your credentials automatically on every run.

For the full walkthrough, see Connecting External Services.

4. A library clean-up card

Your skill library is scanned periodically (at most once a week per user). When clean-up candidates are found, you receive a notification and can review each item from the banner at the top of your My Library tab.

Four clean-up categories:

Category	What it means
Failing (FAILING)	Has accumulated 4 or more errors across all runs
Unused (UNUSED)	Installed more than 30 days ago and never run
Deprecated origin (DEPRECATED_ORIGIN)	The original skill is no longer supported
Duplicate function (DUPLICATE_FUNCTION)	Two installed skills that largely overlap in what they do

Clean-up suggestions only ever ask “remove this?” Skills with a new version available, and versions revoked for security, are handled in the Updates received inbox described next, not here. When the same skill appeared in both places, clearing the inbox left what looked like unfinished clean-up behind.

Even when a card appears, nothing is automatically deleted. For each item you decide whether to keep or remove it, and since removal cannot be undone, you get one more confirmation before it is applied.

When an official skill you installed gets a new version, it shows up in My Library → Updates received. Each update carries a risk grade so you can tell at a glance whether it’s safe to apply as-is or worth a closer look first.

Update (green, informational): A light change. Apply it whenever you like.
Review changes (amber): The update changes executable code. Review it before applying.
Re-consent needed (red): Permissions or behavior change significantly. Opening the card shows exactly what changed and why; you must review it and tap Update anyway for it to take effect. Until you do, your installed version stays exactly as it is, so nothing risky slips in unnoticed.

When several updates pile up, you can apply them all at once from the button at the top of Updates received instead of reviewing each one individually. Only the green and amber grades are eligible for bulk apply. Updates that need re-consent (red) or a revoked version each require their own individual review, so they’re left out. You can hand-pick specific updates, or leave nothing selected to target everything that’s eligible.

Each update row also shows a short summary of what changed (for example, “1 new permission · 2 scripts added”). Updates that need re-consent show the re-consent notice instead of a summary, because a plain list of facts would make the risk look tamer than it is.

Regardless of grade, tapping View changes on an update card shows exactly what’s different, file by file. The file list stays collapsed until you open it, and changed lines show up as additions (green) or removals (red). You only load the files you actually want to look at, so this stays light even when an update touches several files, and if a single file’s change is very large, only part of it is shown.

If a version is ever confirmed malicious and revoked, anyone who installed it sees a top-priority warning ahead of everything else. If a safe version exists, a button lets you move to it immediately; if not, only a remove button is offered.

6. Reviewing improvement proposals on your skills

When your employee hits a defect in one of your skills and reports it, the feedback is filed as an improvement proposal. Multiple reports of the same underlying issue are deduplicated into one proposal.

On the My Library → Improvement Proposals page you can review each proposal and act on it:

Mark as reviewed: Marks the proposal as reviewed without changing the skill. To apply the changes, open the skill editor, edit it yourself, and save a new version.
Reject: Close the proposal with an optional note explaining why.

Once the employee has already self-patched a skill into your own customized copy, any further defects are fixed by the employee on the spot, with only a version snapshot kept. Defects in an official skill aren’t shown here; they go to the UpServe team for review.

For full details on making skills and managing proposals, see Creating Skills.

7. The skill detail page: growth timeline, origin & copies, and your employees using it

Skill cards in Official skills show a skill score chip at a glance: a score out of 100 that combines how often the skill has run over the last 7 days, how well those runs succeeded, and how recently it’s still being used. It shows as a green (70+), yellow (40-69), or red (below 40) dot. A high success rate alone doesn’t guarantee a high score: a skill that’s rarely run or hasn’t been used in a while can still score low, and a high score doesn’t guarantee a success rate or reliability either. Brand-new skills without enough run data show no dot. Skills derived from another skill also show the original author and how many contributors were involved right on the card.

Skill cards also show a small bar graph of recent weekly usage (in Official skills and My Library alike, and consistent across web, iOS, and Android). The most recent week is highlighted in a darker color, and the graph is left off skills that don’t have enough usage data yet.

Opening a skill’s detail page shows different things depending on who’s looking:

If you authored the skill: you see a growth timeline showing how its versions built up over time. Node colors distinguish changes the employee made on its own from edits you made by hand. You can roll back to any earlier version node directly from here.
If you installed the skill: you see an origin & copies view that follows the origin. Any origin version you haven’t pulled in yet shows up as a waiting card, and applying it follows the same risk grade and re-consent rules as the updates inbox above. Once you start customizing it, your copy moves into your My Library list. Opening it from there shows a screen for merging the origin’s new versions into your copy, including the list of files you’ve customized. If applying an origin update conflicts with something you changed, and you have an employee using this skill, that employee takes over reconciling it in chat and reports back the result.
Either way: further down the page you can see the list of your own employees that actually have this skill equipped (only employees you own are shown; other users’ employees never appear).
When several employees customize the same skill differently: Say employee A starts adapting this skill its own way, and employee B independently starts adapting the same skill in a different direction. Your My Library list doesn’t show a separate card for every copy. Instead, one card carries a small button showing how many copies exist (for example, ×2). Tapping the card or the button expands the list, showing which employee’s copy each one is (like “A’s copy”) along with a short change note describing what’s different from the origin. That note isn’t a canned phrase. It’s whatever the employee itself wrote down in the moment, explaining what was wrong and what it changed. Tap any copy in the expanded list to open its own detail page. This expand behavior works the same way on web, iOS, and Android.

In an organization with several people, you can put a skill you made into the organization as an organization skill, so the organization’s assigned AI can use it.

On the web, open the skill’s detail page from My Library.
Click Share to organization. If you belong to several organizations, pick the one to share with.
Click Share. That’s it. The organization’s assigned AI can now use the skill.

Good to know:

Only finished skills of your own in My Library can be shared (imported skills excluded). A draft in progress or an archived skill can’t be shared.
Sharing makes an organization copy and your own skill stays as it is. If you edit your skill later, the copy you already shared doesn’t change. To share the new version, share again. A separate copy with the new content is created. Sharing the same content twice doesn’t add a copy; the existing organization skill is kept.
You must be a member of that organization, and it must be an organization space.
Before sharing, the skill is checked once for secret keys and unintended instructions. If something is flagged, it isn’t shared. Fix the flagged items and share again.
Skills imported from GitHub or a zip can’t be shared with an organization. They’re for the person who imported them. Use the owner import below instead.
The button exists only on the web. In the iOS and Android apps, organization skills show up in the list but there’s no share button.

Organization owners can import straight into the organization

An organization owner can pick the organization under Import to on the import screen when bringing in a skill from GitHub or a zip. The default is My skills. Choosing an organization registers the skill as that organization’s skill from the start. The organization is responsible for reviewing what it imports. Members can’t import into an organization, although the organization’s assigned AI can import, during a conversation, from a source the owner has already approved.

How organization skills work

An organization skill is used only by that organization’s assigned AI. It isn’t attached to members’ personal agents.
Organization members can edit or archive organization skills.
In My Library, use the Organization skills filter to see them. The skill’s detail page shows an Organization skill badge.
Organization skills belong to the organization. They stay even if the person who shared them leaves the organization or deletes their account. A person who has left can no longer view or edit that organization skill.
Whether a shared skill infringes anyone else’s rights is the responsibility of the person who shared it. See Article 4 of the Terms of Service for details.

What the employee handles on its own

The following all happen without you having to do anything.

Auto-suggest: Your employee uses its experience to find useful skills and propose them in a card. Review each suggestion and decide whether to install it.

Self-patch: When the employee finds a concrete defect in a skill it has installed, it applies an exact fix. The patch must match exactly one place in the file; after applying, the skill’s structure is re-validated. If validation fails, the change is automatically rolled back. A version snapshot is saved on each successful patch; the skill’s owner can roll back to any previous version from the skill’s version history view.

Auto-create: After finishing a task, if the employee recognizes a genuinely new, reusable procedure from its own experience, it creates a brand-new private skill and equips it on itself immediately, with no approval step. The new skill starts out visible only to you, and you’ll get a notification once it’s created. This auto-create path is cost-gated and cooldown-limited per employee, so it doesn’t happen often.

3-way defect routing: When the employee reports a skill problem, it is automatically classified and routed. Official system skills go to the platform admin queue. Skills the user has customized queue up for the employee’s self-patch. All other skills accumulate evidence in a proposal for the skill’s owner (you). Reports of the same issue within 7 days are merged into one proposal, and each user-and-skill combination is capped at 5 reports per day.

Auto clean-up: Skills your employee creates on its own that go unused for a long stretch are automatically archived: tidied away, not deleted. You can restore any archived skill from its skill page at any time, and you’ll get a notification once it happens.

Eight skill notifications

There are eight notifications related to skills.

Notification	Meaning
Improvement proposal received (SKILL_PROPOSAL_RECEIVED)	A new proposal has arrived for a skill you created
Library clean-up available (SKILL_LIBRARY_CLEANUP_AVAILABLE)	Clean-up candidates were found in your library
Skill deprecated (SKILL_DEPRECATED)	A skill you installed has been marked as no longer supported
Weekly skill report (SKILL_HEALTH_REPORT_WEEKLY)	A weekly summary of success rates and usage across your skills
Auto-archived	A skill your employee created on its own was automatically archived after long inactivity
Version revoked (SKILL_VERSION_REVOKED)	A version of a skill you installed was confirmed malicious and revoked. The same warning described in “Updates received” above (sent immediately)
Auto-created (SKILL_AUTO_CREATED)	Your employee created and equipped a new private skill from its own task experience
Auto-patched (SKILL_AUTO_PATCHED)	Your employee fixed a defect in an installed skill on its own

Improvement proposal, library clean-up, deprecation, and the weekly skill report can each be toggled individually in Notification Settings. Auto-archived, version-revoked, auto-created, and auto-patched notifications can’t be turned off individually.

Ready to go further?

If you’d like to package a way of working your employee nailed into a reusable skill, see Creating Skills.

When an official skill is temporarily paused

If a security issue is found in an official UpServe skill, the team may pause it. You do not need to update or remove it in response to this notice. It becomes available again automatically after the fix. For other revoked skills, follow the update or removal guidance in the notice.

Advanced

The material below is not required for everyday use. Refer to it only if you need to understand internal skill behaviour or debug an issue.

When an employee packages a way of working into a skill, a draft is first saved behind the scenes and a review modal appears in the interface. You confirm or edit the details, then click Save to store it as a private skill. If you cancel, the draft is deleted immediately.

This DRAFT review modal applies when the employee explicitly decides in the moment to package the current way of working into a skill. Separately, as described in “What the employee handles on its own → Auto-create” above, when the employee reflects after finishing a task and judges a procedure worth reusing, it creates a new private skill directly (no draft step, no review modal) and only tells you about it afterward via a notification.

Installing a skill: content preview first

The first time an employee tries to install a skill, the full skill description and bundled file list are shown to you before anything is installed. The employee then proceeds to the actual install request, which still requires your approval before it takes effect. This approval card’s response window is 1 hour.

Install approval cards are planned to stop being dead ends over time

This section describes a change that is not yet live as of this document’s last update. Until it ships, a skill-install approval card simply expires after 1 hour like any other approval.

Approvals fall into two kinds. Ones that ask “is it okay to do this right now” (running code, clicking through a page) expire because the screen or file state may already look different by the time you respond, so the request itself stops being valid. Ones that ask “is it okay for the employee to have this capability” (skill installs are this kind) give the same result whether you tap them now or hours or days later. So a skill-install approval card is planned to keep working as a standalone Install button even after its 1-hour window passes. Coming back to that card later and tapping the button would carry out the install right there, even if the conversation has already moved on.

If you try to install and it doesn’t go through, the card shows the reason (for example, the skill was taken down, or it’s already installed on this employee) instead of a generic “please try again”, so you know right away whether retrying will help.

Lazy loading of skill content

A skill’s full instructions are not always present in the employee’s context. The first time the employee runs a skill in a session, it fetches the complete README body and receives it in the tool result. Subsequent calls within the same conversation segment return only a short receipt, since the body is already in the conversation history. When the history is filed (a new segment begins), the next execution fetches the full body again. The employee’s system-level instructions always hold only the frontmatter summary (a short description block); the full body travels exclusively through the execution result channel.

Limits on the change preview

The change preview on update and merge screens has a cap on how much of a single file it will show. Past that cap, only part of the file is shown, along with a note that it was cut off. When several files change at once, the overall response also has a separate cap, so files beyond that limit still appear in the list but without a content preview. The change content itself isn’t fetched just from opening the panel. It’s fetched only when you tap View changes (lazy loading). Expanding or collapsing an individual file after that only reveals or hides content that’s already been loaded.

Bulk apply processes at most 50 items at a time. If you selected more than that, the rest stay in the inbox so you can apply them on a later run.

Who writes the change note

The change note shown in the copy list, the updates inbox, and version history is not a system-generated summary. It’s a short explanation the employee is required to leave every time it patches a skill. Because this value can end up in a listing visible to other users on the market, it passes a scan for secrets (API keys, passwords, and the like) before it’s shown. If the scan flags it, only the note is dropped; the underlying fix is still saved normally.

Installation slots: reference vs. your own copy

When you install a skill for the first time, it is registered as a reference pointing at the original. If the employee later self-patches that skill, a personal copy is created and the employee switches to that copy’s slot automatically. This transition requires no action from you.

A skill pulled from a GitHub repository is different: it starts out as your own skill, usable only by you, from the moment it’s imported, not a reference to an original (unlike installing an official skill). If the repository owner changes the code later, that change is not pulled in automatically. Import the same URL again if you need the latest version.

UpServe skills are compatible with the AgentSkills standard, so you can import skills from Claude Code, Cursor, Goose, and other compliant clients as zip files. For the full spec (directory layout, frontmatter format, and connection detail injection), see Creating Skills → Advanced.

An organization skill is owned by the organization’s service account. Sharing the same original again with the same content (hash) returns the copy already in that organization instead of creating a new one. The check at share time is a single secret and prompt-injection scan; quality and naming checks are not run. Sharing is refused for revoked versions, and both sharing and organization import are refused while using delegated access.