Skills
Merge’s Skills Library is the governed place your organization publishes, permissions, and serves the skills its agents run.
A skill is a Markdown instruction package that teaches an agent how to do a repeatable piece of work: close the books, qualify an inbound lead, write up a customer call. It is a plain SKILL.md file, plus any reference files it needs, built to the open SKILL.md format.
Admins publish a skill once, and every AI client your employees already use reads it from the same place: Claude, ChatGPT, Cursor, Claude Code, and anything else pointed at your MCP endpoint. Use it when the same task run by ten people through AI produces ten different results, and you want one approved way to do the work.
If tools answer “what can this agent call,” skills answer “how should it do the work.”
Skills are part of Agent Handler for Employees, the setup where your own people reach company tools from their own AI clients. Agents you embed in your product, covered by Building an agent, do not retrieve skills today.
Availability
Skills is available to every organization, and you do not need to turn anything on to use it. If you would rather your members not publish or retrieve skills at all, ask your Merge contact to switch it off for your org.
What a skill is made of
Every skill is a bundle with a SKILL.md at its root. That file carries YAML frontmatter (a name and a description) and a Markdown body holding the actual instructions.
A skill can also bundle supporting files: longer reference material, example payloads, a checklist an agent only needs for one branch of the work. Agents load SKILL.md first and pull a bundled file only when the instructions send them to it, so the entry point stays small and the detail stays available. See Writing a skill for the full format and limits.
How skills relate to tools and connectors
A skill is advisory. It tells an agent what to do and in what order; it grants nothing.
Agent Handler executes the tool calls, and every call is still checked against the user’s own permissions and the Tool Pack in play. A skill that says “update the closed-won opportunity in Salesforce” fails for a user with read-only Salesforce access, exactly as a direct request would. Skills cannot widen access, so publishing one is not a permission decision.
Reach is the one thing a skill does control: who can read it. A skill is published to the whole organization, to specific provisioning groups, or to its author alone, and list_skills is filtered per user accordingly. That decides who sees the instructions, never what they are allowed to call. See Who a skill reaches.
A skill’s frontmatter can name the connectors it expects (referenced_connectors), which is how the dashboard shows the connectors a skill leans on. That list is a declaration for humans reading the card, not a grant and not a precondition.
Your library, and where it comes from
Your library is not one list someone maintains. It is the set of skills that reach you, assembled from three places, and the dashboard labels each skill with its origin.
- Skills you authored. Your own, whether you published them to yourself or wrote them for the org.
- Skills your org published. What your admins published organization-wide, or to a group you belong to.
- Skills your admin enabled from Browse. Merge-maintained skills an admin turned on for the org.
That last one is what Browse is: a catalog of ready-made skills you can read through and turn on, rather than write from scratch. It holds the ones Merge writes by hand and the ones Merge generates, and nothing in it reaches your members until an admin enables it. A growing catalog never quietly expands what your agents read.
Two people in the same org can therefore have different libraries, because group-scoped and personal skills reach different people. See Who a skill reaches.
What gets recorded
Skills writes to the same audit trail as the rest of Agent Handler. Publishing, drafting, submitting, approving, rejecting, unpublishing, republishing, enabling a catalog skill, disabling one, and every agent retrieval each land as an event attributed to the user who caused it. So “which skill did this agent read before it made that call” is answerable after the fact, on the Audit Trail.
Getting started
Publishing your first skill takes one file. Write a SKILL.md, upload it on the Skills page, and it is available to your agents through list_skills and retrieve_skill on the next call.
Next: Writing a skill