@runtypelabs/cli 1.8.2 → 1.8.4

package/README.md

@@ -62,6 +62,14 @@ runtype marathon agent_abc123 --goal "Write integration tests for the API" \
 # Override the agent's configured model
 runtype marathon "Code Builder" --goal "Build a calculator" --model claude-sonnet-4-5
 
+# Enable built-in tools for the task (web search, scraping, image generation, etc.)
+runtype marathon "Code Builder" --goal "Search the web and summarize" --tools exa
+runtype marathon "Code Builder" -g "Find docs and scrape them" --tools exa firecrawl
+
+# Combine tools with other options
+runtype marathon "Code Builder" --goal "Research and generate images" \
+  --tools exa dalle --max-sessions 5
+
 # Enable sandboxed code execution tooling for the task (QuickJS or Daytona)
 runtype marathon "Code Builder" --goal "Compute totals from this dataset" --sandbox quickjs
 runtype marathon "Code Builder" -g "Generate a script and run it" --sandbox daytona
@@ -124,6 +132,42 @@ runtype marathon "Code Builder" --goal "Fix the bug" \
 runtype marathon "Code Builder" --goal "Fix the bug" --tool-context full-inline
 ```
 
+#### Built-in Tools
+
+The `--tools` (or `-t`) flag enables built-in platform tools during agent execution. Tools are validated at startup against the built-in tools registry, and compatibility is checked against all models used in the task (including planning and execution phase models).
+
+```bash
+# Enable a single tool
+runtype marathon "Code Builder" --goal "Search the web" --tools exa
+
+# Enable multiple tools (space-separated)
+runtype marathon "Code Builder" --goal "Research and generate" --tools exa firecrawl dalle
+
+# Short flag
+runtype marathon "Code Builder" -g "Scrape docs" -t firecrawl
+```
+
+**Available tools:**
+
+| Tool ID          | Description                                 |
+| ---------------- | ------------------------------------------- |
+| `exa`            | Semantic web search                         |
+| `firecrawl`      | Web scraping and content extraction         |
+| `dalle`          | DALL-E image generation                     |
+| `elevenlabs-tts` | Text-to-speech via ElevenLabs               |
+| `elevenlabs-stt` | Speech-to-text transcription via ElevenLabs |
+
+**Not available via CLI:**
+
+- `openai_web_search`, `anthropic_web_search`, `anthropic_web_fetch` — provider-native tools handled automatically by compatible models
+- `vector-search` — requires additional configuration (use the dashboard instead)
+
+**Validation behavior:**
+
+- Unknown tool IDs produce an error listing all available tools
+- Tools incompatible with the selected model(s) are rejected with a specific error
+- All validation errors are reported together so you can fix them in one pass
+
 ## Configuration
 
 ```bash