fix(mcp): SKILL.md documents the business-action tools (#2714 Phase 0)

.changeset/skill-document-action-tools.md

@@ -0,0 +1,17 @@
+---
+'@objectstack/mcp': patch
+---
+
+fix(mcp): the generated SKILL.md now documents the business-action tools
+
+`renderSkillMarkdown()` listed only the 7 object-CRUD tools; the MCP surface
+exposes 9 — `list_actions` / `run_action` (business actions) were missing, so
+agents installing the skill never learned they can run approvals, conversions,
+or flow triggers directly. The skill now covers the full native tool surface
+and teaches action preference: when `list_actions` offers a matching action,
+call it instead of hand-editing the records it would have touched (actions
+carry the app's validation and side effects), confirming destructive or
+confirmation-flagged actions with the user first.
+
+Prerequisite for the distribution shells (#2714 Phase 0): every shell repo
+copies this rendered content, so the gap had to close before fan-out.

packages/mcp/src/skill.test.ts

@@ -51,7 +51,7 @@ describe('renderSkillMarkdown', () => {
     expect(md).toContain('Authorization: ApiKey');
   });
 
-  it('lists the object-CRUD tools and a discover-first instruction', () => {
+  it('lists the full native tool surface (CRUD + business actions) and a discover-first instruction', () => {
     const md = renderSkillMarkdown();
     for (const tool of [
       'list_objects',
@@ -61,12 +61,20 @@ describe('renderSkillMarkdown', () => {
       'create_record',
       'update_record',
       'delete_record',
+      'list_actions',
+      'run_action',
     ]) {
       expect(md).toContain(tool);
     }
     expect(md.toLowerCase()).toContain('discover');
   });
 
+  it('teaches action preference — run a matching business action instead of hand-editing records', () => {
+    const md = renderSkillMarkdown();
+    expect(md).toContain('Prefer actions over hand-edits');
+    expect(md).toContain('run_action({ actionName, objectName?, recordId?, params? })');
+  });
+
   it('is generic — it does not enumerate any concrete schema', () => {
     // The skill must not bake in tenant object/field names; it points to live
     // discovery instead. Sanity: no stray "objectName: <something concrete>".

packages/mcp/src/skill.ts

@@ -58,18 +58,21 @@ description: ${OBJECTSTACK_SKILL_DESCRIPTION}
 
 # ObjectStack
 
-${intro} An ObjectStack environment exposes its data **objects** (tables) as
-tools over the Model Context Protocol (MCP). Every operation runs **as you** —
+${intro} An ObjectStack environment exposes its data **objects** (tables) and
+its business **actions** (registered app logic: approvals, conversions, flow
+triggers) as tools over the Model Context Protocol (MCP). Every operation runs **as you** —
 under your account's permissions and row-level security — so you may see a
 subset of rows, or get a permission error on a write. That is expected
 governance, not a failure.
 
 ## When to use
 
 Use these tools whenever the user wants to **inspect or change data** in their
-ObjectStack app: look up records, filter/report, create or update entries, or
-clean up data. Prefer these tools over guessing — the environment is the source
-of truth.
+ObjectStack app — look up records, filter/report, create or update entries,
+clean up data — or **run a business action** the app defines (approve a
+request, convert a lead, kick off a flow). Prefer these tools over guessing —
+the environment is the source of truth, and an action is always better than
+hand-editing the records it would have touched.
 
 ## Connect
 
@@ -112,6 +115,8 @@ always current even as the app evolves:
 1. \`list_objects\` — see what objects exist.
 2. \`describe_object({ objectName })\` — get an object's fields (name, type,
    required) before querying or writing it.
+3. \`list_actions\` — see what business actions you may run (each entry
+   includes its parameters and whether it needs a \`recordId\`).
 
 Always discover the relevant object's shape before constructing a filter or a
 create/update payload.
@@ -128,6 +133,14 @@ create/update payload.
 - **update_record({ objectName, recordId, data })** — change fields on a record.
 - **delete_record({ objectName, recordId })** — delete a record (destructive —
   confirm with the user first).
+- **list_actions()** — list the business actions you are permitted to run, with
+  each action's declared parameters, whether it operates on a record, and
+  whether it is flagged destructive.
+- **run_action({ actionName, objectName?, recordId?, params? })** — invoke a
+  business action by name. This executes the app's registered logic (can
+  mutate data or trigger flows) under your permissions and RLS. Pass
+  \`recordId\` for record-scoped actions and \`params\` for declared inputs;
+  \`objectName\` only disambiguates a name shared by multiple objects.
 
 ## Conventions & gotchas
 
@@ -139,13 +152,20 @@ create/update payload.
 - **Writes are real and immediate.** There is no implicit dry-run. Confirm
   destructive actions (\`delete_record\`, bulk updates) with the user.
 - **Page large reads.** Use \`limit\`/\`offset\` rather than asking for everything.
+- **Prefer actions over hand-edits.** When \`list_actions\` offers an action for
+  the task (approve, convert, close, …), call it instead of updating the
+  records yourself — actions carry the app's validation and side effects.
+  Confirm first when an action is record-destructive or flagged for
+  confirmation.
 
 ## Recommended workflow
 
-1. \`list_objects\` to orient.
+1. \`list_objects\` (and \`list_actions\` when the task sounds like a business
+   operation) to orient.
 2. \`describe_object\` on the target object.
 3. \`query_records\` to read / verify current state.
-4. \`create_record\` / \`update_record\` / \`delete_record\` to make changes,
-   confirming destructive steps with the user.
+4. \`run_action\` when a matching business action exists; otherwise
+   \`create_record\` / \`update_record\` / \`delete_record\` — confirming
+   destructive steps with the user.
 `;
 }