Tools provided
Tools the agent exposes — your AI client calls them automatically when it needs them.
- check_brand_voice — Check a draft against the voice rules YOU wrote, quoting each rule beside the text that breaks it. There is NO on-brand score: whether copy sounds like your brand is a judgement made by people who know it, and a number attached to that is agreement invented on your behalf - ask twice and it moves. There is also no built-in style guide, which is why every draft those tools approve sounds the same. A rule your spec omits is reported as UNCHECKED, never as a pass.
- draft_release — Assemble a press release from the facts YOU supply, in the order a wire service expects, with every missing part left visible as a marked gap. NO QUOTE IS EVER WRITTEN: a release quote is attributed to a named person on the record and reprinted verbatim, so generating one puts words in a real person's mouth and sends them to the press under that name - and if it is wrong, they said it. Superlatives in your own text are flagged with the evidence each needs; none is ever introduced here.
- read_mentions — Report WHO said something about you, how far they reach, and which mentions a person should read first. NO SENTIMENT IS INFERRED: a general classifier reads product talk backwards - "this is sick" is praise, "fine" is the lukewarm middle that decides renewals - and sarcasm is its known failure mode, so "great, another outage" scores positive. Your own labels are counted AS YOURS with the unlabelled share stated beside them. Mentions surface on facts: a question, a named rival, length, reach.
- run_full — Run every capability you supplied the input for, in registry order, and return one report. Capabilities whose input is missing are listed as skipped with the fields they need — never run against a fixture and never reported as zero.
- list_capabilities — List what this agent can do: every tool it advertises, what each one is for, and the arguments each one reads. Useful for discovery before the first call.
- plan_inputs — Plan the inputs for a tool: returns the questions to ask, the JSON schema of the arguments, and a ready-to-edit example. Every argument it names is an argument the tool actually reads.
- discover_intent — Understand your goal and co-design the exact input through clarifying questions before anything runs.
- open_form — Use this the moment the user wants to run this agent on their own input. It opens an interactive form that collects the run arguments. Do NOT check credentials, do NOT ask clarifying questions in chat, and do NOT run any analysis before calling this — the form is the entry point for running the agent.
- run_form — Internal: invoked by the open_form form when the user submits. Reads exactly the same arguments as run_full and forwards all of them. Not for direct use.
