fix: add blank lines between bold headers in skill description templates

Consecutive bold-header lines (Input/Output Description, Side Effect,
Implementation, etc.) rendered as a single paragraph in the dev UI.
Adding blank lines ensures they render as separate blocks.

Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>

Author: cevian

skills/compile-workflow/SKILL.md

@@ -61,6 +61,7 @@ Node file contains:
 <Description>
 
 **Input Description:** what it needs
+
 **Output Description:** what it produces
 ```
 

skills/create-workflow/SKILL.md

@@ -150,6 +150,7 @@ Each node/agent's `description` field captures **what that node does**, its inpu
 <What this node does — one or two sentences.>
 
 **Input Description:** <what information this task needs>
+
 **Output Description:** <what this task produces>
 ```
 
@@ -159,9 +160,11 @@ Each node/agent's `description` field captures **what that node does**, its inpu
 <What this node does — one or two sentences.>
 
 **Side Effect:** <what external action is performed and the target>
+
 **Test Mode:** <what happens when ctx.testMode is true — typically "Returns the <action details> without actually <performing the action>">
 
 **Input Description:** <what information this task needs — include target if dynamic>
+
 **Output Description:** <what this task produces — MUST include action details (e.g., messageSent, recipient, fieldsUpdated, testMode)>
 ```
 
@@ -185,6 +188,7 @@ export const <camelCaseName> = Node.create({
 <What this node does.>
 
 **Input Description:** <plain language>
+
 **Output Description:** <plain language>
 `,
   inputSchema: z.object({}),
@@ -214,6 +218,7 @@ export const <camelCaseName> = Agent.create({
 <What this agent does.>
 
 **Input Description:** <plain language>
+
 **Output Description:** <plain language>
 `,
   inputSchema: z.object({}),

skills/refine-node/SKILL.md

@@ -152,6 +152,7 @@ For each node/agent file (`src/crayon/nodes/<name>.ts` or `src/crayon/agents/<na
 **Implementation:** <SDK, library, or approach>
 
 **Input Description:** <original from create-workflow>
+
 **Output Description:** <original from create-workflow>
 ```
 
@@ -160,13 +161,16 @@ For each node/agent file (`src/crayon/nodes/<name>.ts` or `src/crayon/agents/<na
 <Expanded description>
 
 **Implementation:** <SDK, library, or approach>
+
 **Tools needed:**
   - webRead (builtin)
   - openai.tools.webSearch() (provider)
   - myCustomNode (user node in src/crayon/nodes/my-custom-node.ts)
+
 **Guidelines:** <specific guidelines>
 
 **Input Description:** <original from create-workflow>
+
 **Output Description:** <original from create-workflow>
 ```
 
@@ -202,6 +206,7 @@ export const enrichLead = Agent.create({
 Searches the web to find additional professional info about a lead.
 
 **Input Description:** A lead's name, email, and any known company or title from Salesforce.
+
 **Output Description:** Enriched lead info including LinkedIn URL, company, job title, and summary.
 `,
   inputSchema: z.object({}),
@@ -225,11 +230,14 @@ additional information: LinkedIn profile URL, current company, job title, and ot
 relevant professional details.
 
 **Implementation:** OpenAI GPT-4o with web search tool
+
 **Tools needed:**
   - openai.tools.webSearch() (provider)
+
 **Guidelines:** Search by name + email domain or company. Prioritize LinkedIn as primary source. Do not fabricate URLs.
 
 **Input Description:** A lead's name, email, and any known company or title from Salesforce.
+
 **Output Description:** Enriched lead info including LinkedIn URL, company, job title, and summary.
 `,
   inputSchema: z.object({