Tools provided
Tools the agent exposes — your AI client calls them automatically when it needs them.
- run_full — Orchestrated path: repurpose one provided source into the full channel package — LinkedIn post, X thread, newsletter section, video script and carousel copy — with deterministic counts throughout.
- analyze — Run pure deterministic repurposing over a provided source with no model call and return the per-channel outputs with their unit counts.
- extract_key_points — Extract the key points from the provided source — the step before any channel copy is written, and a useful check on what the agent thinks the piece is about.
- write_linkedin_post — Write just the LinkedIn post — hook, key-point bullets and call to action — with its exact character count and limit flag.
- write_x_thread — Write just the X thread — a hook tweet, one tweet per key point and a closing CTA — with its exact tweet count.
- write_newsletter — Write just the newsletter section — headline, intro, takeaways and link — with its word count.
- write_video_script — Write just the sixty-second video script, laid out beat by beat from hook to outro.
- write_carousel — Write just the carousel slide copy, one slide per key point, with its slide count.
- list_capabilities — List the agent's static capabilities: the available tools and the supported channels. Useful for discovery.
- plan_inputs — Plan the inputs for a tool: returns the clarifying questions, the JSON schema and a ready-to-edit example.
- discover_intent — Understand your goal and co-design the exact input through clarifying questions before running.
- open_form — Opens the guided form to collect the source content, the channels you want and the tone. The entry point for running the agent on your own material.
- run_form — Internal: invoked by the guided form when the user submits. Not for direct use.
