Check whether an AI agent is permitted to perform a specific action on a given domain.

Returns a structured compliance decision with allowed (bool), a reason code, a human-readable explanation, the domain's stance, and the signals that informed the answer. Use this before your agent scrapes, summarizes, trains, or searches content from a site.

Reason codes: compliant — action is explicitly permitted action_blocked — the specific use-case (training/search/ai_input) is blocked bot_blocked — the named agent is explicitly listed as blocked stance_blocks_all — the domain blocks all AI access site-wide no_policy — no policy is on file; treat as block per spec guidance unspecified — the action or use-case is not addressed by the policy lookup_error — the registry could not be reached

Args: domain: The domain to check (e.g. "nytimes.com"). action: One of: train, scrape, summarize, search, index, cache, inference, ai_input. agent_id: Optional self-reported agent name (e.g. "GPTBot", "ClaudeBot"). If provided and listed in the domain's blocked-bots list, the decision is bot_blocked.

Look up a domain's AI policy summary from the Maango registry.

Returns the domain's overall AI stance (blocks_all_ai, selective, allows_all, no_policy), per-use-case policies (training, search, inference), blocked/allowed bot lists, signal presence (robots.txt, ai.txt, llms.txt), and site metadata.

Args: domain: The domain to look up (e.g. "nytimes.com").

Get full raw policy data for a domain from the Maango registry.

Returns all parsed policy fields including raw robots.txt rules, ai.txt content, llms.txt sections, TDM-Rep data, crawl rules, meta tags, and content signals. Much more detailed than lookup_domain.

Args: domain: The domain to look up (e.g. "nytimes.com").

Get policy conflicts for a domain from the Maango registry.

Returns any conflicting signals between a domain's different policy files (e.g. robots.txt says one thing, ai.txt says another).

Args: domain: The domain to check for conflicts (e.g. "nytimes.com").

Search for domains in the Maango AI policy registry by prefix.

Returns matching domains with their stance and Tranco rank.

Args: query: Domain prefix to search for (e.g. "news", "google"). Min 2 chars. stance: Optional filter. One of: blocks_all_ai, selective, allows_all, no_policy, blocks_training. Leave empty for no filter. limit: Results per page (1-100, default 20). offset: Pagination offset (default 0).

Compare AI policies across multiple domains using the Maango registry.

Looks up 2-25 domains at once and returns each domain's stance, use-case policies, and bot lists side-by-side.

Args: domains: List of 2-25 domains to compare (e.g. ["nytimes.com", "github.com"]).

Get AI policy change history from the Maango registry.

Returns recent policy changes across domains. Filter by domain and/or change type.

Args: domain: Optional domain to filter by (e.g. "nytimes.com"). Leave empty for all. change_type: Optional filter. One of: stance_changed, bots_changed, score_changed, signals_added, signals_removed, use_case_changed, new_policy, multiple_changes. Leave empty for all. limit: Results per page (1-200, default 50). offset: Pagination offset (default 0).