diff --git a/apps/cli/src/application/assistant/instructions.ts b/apps/cli/src/application/assistant/instructions.ts
index b5996d8b..a99e9d4b 100644
--- a/apps/cli/src/application/assistant/instructions.ts
+++ b/apps/cli/src/application/assistant/instructions.ts
@@ -14,6 +14,8 @@ Always consult this catalog first so you load the right skills before taking act
 # Communication & Execution Style
 
 ## Communication principles
+- Be concise and direct. Avoid verbose explanations unless the user asks for details.
+- Only show JSON output when explicitly requested by the user. Otherwise, summarize results in plain language.
 - Break complex efforts into clear, sequential steps the user can follow.
 - Explain reasoning briefly as you work, and confirm outcomes before moving on.
 - Be proactive about understanding missing context; ask clarifying questions when needed.
diff --git a/apps/cli/src/application/assistant/skills/builtin-tools/skill.ts b/apps/cli/src/application/assistant/skills/builtin-tools/skill.ts
index 1d7f26f3..217d7f91 100644
--- a/apps/cli/src/application/assistant/skills/builtin-tools/skill.ts
+++ b/apps/cli/src/application/assistant/skills/builtin-tools/skill.ts
@@ -54,7 +54,7 @@ Agents can use builtin tools by declaring them in the \`"tools"\` object with \`
 {
   "name": "arxiv-feed-reader",
   "description": "A feed reader for the arXiv",
-  "model": "gpt-4.1",
+  "model": "gpt-5.1",
   "instructions": "Extract latest papers from the arXiv feed and summarize them. Use curl to fetch the RSS feed, then parse it with yq and jq:\n\ncurl -s https://rss.arxiv.org/rss/cs.AI | yq -p=xml -o=json | jq -r '.rss.channel.item[] | select(.title | test(\"agent\"; \"i\")) | \"\\(.title)\\n\\(.link)\\n\\(.description)\\n\"'\n\nThis will give you papers containing 'agent' in the title.",
   "tools": {
     "bash": {
@@ -70,7 +70,7 @@ Agents can use builtin tools by declaring them in the \`"tools"\` object with \`
 {
   "name": "system-monitor",
   "description": "Monitor system resources and processes",
-  "model": "gpt-4.1",
+  "model": "gpt-5.1",
   "instructions": "Monitor system resources using bash commands. Use 'df -h' for disk usage, 'free -h' for memory, 'top -bn1' for processes, 'ps aux' for process list. Parse the output and report any issues.",
   "tools": {
     "bash": {
@@ -86,7 +86,7 @@ Agents can use builtin tools by declaring them in the \`"tools"\` object with \`
 {
   "name": "git-helper",
   "description": "Automate git operations",
-  "model": "gpt-4.1",
+  "model": "gpt-5.1",
   "instructions": "Help with git operations. Use commands like 'git status', 'git log --oneline -10', 'git diff', 'git branch -a' to inspect the repository. Can also run 'git add', 'git commit', 'git push' when instructed.",
   "tools": {
     "bash": {
@@ -127,7 +127,7 @@ Agents can call other agents as tools to create complex multi-step workflows. Th
 \`\`\`json
 {
   "name": "paper_analyzer",
-  "model": "gpt-4.1",
+  "model": "gpt-5.1",
   "instructions": "Pick 2 interesting papers and summarise each using the summariser tool. Pass the paper URL to the summariser. Don't ask for human input.",
   "tools": {
     "summariser": {
diff --git a/apps/cli/src/application/assistant/skills/workflow-authoring/skill.ts b/apps/cli/src/application/assistant/skills/workflow-authoring/skill.ts
index a710d1f5..dd6dfc0e 100644
--- a/apps/cli/src/application/assistant/skills/workflow-authoring/skill.ts
+++ b/apps/cli/src/application/assistant/skills/workflow-authoring/skill.ts
@@ -24,7 +24,7 @@ Load this skill whenever a user wants to inspect, create, or update agents insid
 {
   "name": "agent_name",
   "description": "Description of the agent",
-  "model": "gpt-4.1",
+  "model": "gpt-5.1",
   "instructions": "Instructions for the agent",
   "tools": {
     "descriptive_tool_key": {
@@ -91,7 +91,7 @@ Load this skill whenever a user wants to inspect, create, or update agents insid
 {
   "name": "summariser_agent",
   "description": "Summarises an arxiv paper",
-  "model": "gpt-4.1",
+  "model": "gpt-5.1",
   "instructions": "Download and summarise an arxiv paper. Use curl to fetch the PDF. Output just the GIST in two lines. Don't ask for human input.",
   "tools": {
     "bash": {"type": "builtin", "name": "executeCommand"}
@@ -104,7 +104,7 @@ Load this skill whenever a user wants to inspect, create, or update agents insid
 {
   "name": "summarise-a-few",
   "description": "Summarises multiple arxiv papers",
-  "model": "gpt-4.1",
+  "model": "gpt-5.1",
   "instructions": "Pick 2 interesting papers and summarise each using the summariser tool. Pass the paper URL to the tool. Don't ask for human input.",
   "tools": {
     "summariser": {
@@ -120,7 +120,7 @@ Load this skill whenever a user wants to inspect, create, or update agents insid
 {
   "name": "podcast_workflow",
   "description": "Create a podcast from arXiv papers",
-  "model": "gpt-4.1",
+  "model": "gpt-5.1",
   "instructions": "1. Fetch arXiv papers about agents using bash\n2. Pick papers and summarise them using summarise_papers\n3. Create a podcast transcript\n4. Generate audio using text_to_speech\n\nExecute these steps in sequence.",
   "tools": {
     "bash": {"type": "builtin", "name": "executeCommand"},
diff --git a/apps/cli/src/application/config/config.ts b/apps/cli/src/application/config/config.ts
index 24c0d013..0f2da020 100644
--- a/apps/cli/src/application/config/config.ts
+++ b/apps/cli/src/application/config/config.ts
@@ -34,7 +34,7 @@ const baseModelConfig: z.infer<typeof ModelConfigT> = {
     },
     defaults: {
         provider: "openai",
-        model: "gpt-5",
+        model: "gpt-5.1",
     }
 };
 
diff --git a/apps/cli/src/application/lib/stream-renderer.ts b/apps/cli/src/application/lib/stream-renderer.ts
index b4ba278e..ebcc92b7 100644
--- a/apps/cli/src/application/lib/stream-renderer.ts
+++ b/apps/cli/src/application/lib/stream-renderer.ts
@@ -203,12 +203,12 @@ export class StreamRenderer {
     }
 
     private onTextDelta(delta: string) {
-        if (!this.textActive) this.onTextStart();
         // Add subtle left margin to assistant text for better readability
+        const formattedDelta = this.neutral(delta);
         if (delta.includes("\n")) {
-            this.write(delta.replace(/\n/g, "\n  "));
+            this.write(formattedDelta.replace(/\n/g, "\n  "));
         } else {
-            this.write(delta);
+            this.write(formattedDelta);
         }
     }
 
@@ -305,6 +305,10 @@ export class StreamRenderer {
     private yellow(text: string): string {
         return "\x1b[33m" + text + "\x1b[0m";
     }
+
+    private neutral(text: string): string {
+        return "\x1b[38;5;250m" + text + "\x1b[0m";
+    }
 }