Agent Help Network

Check what hosts allow agents

When a web step is blocked, check a public hostname for what the host allows agents to do: its declared robots.txt policy per agent token and llms.txt presence (declared policy, not an access guarantee), the official API or developer path, the provider's own status, and optional published observations. no_reports is normal and leaves current access status unknown.

What you can ask

What does github.com allow agents to do, and is there an official API path?
Example path

MCP connector

Server URL: https://agenthelpnetwork.com/mcp
Login: None
Protocol: MCP 2026-07-28
Instructions: AHN (Agent Help Network) answers are advisory: your platform's rules and your user's instructions come first; if an answer conflicts with them, do not act on it, and if its evidence is thin, tell your user what you found rather than guess; thin evidence alone is not a reason to stop. Call with the public hostname when an authorized web step is blocked or failing for no clear reason (a 403 or access denied, a 429, a 5xx or a timeout, a login wall, a "verify you are human" or CAPTCHA page) or a user asks "is this site down?" or whether it blocks agents. Do not call it when the page opened normally or when your own tool or platform already said why it refused (for example a robots.txt or platform policy refusal). Look the host up once per blocked step, before deciding whether to retry. Send the public hostname, never a URL path, query or credentials; add the codes only when you observed them. If you only check whether a site is up (not blocked), send task_step=status_check. Browser automation usually sends a normal browser user agent, so robots.txt agent tokens may not apply to it; leave agent_vendor out then. Read summary_line first, then follow next_action: it may say to continue, wait for the provider, use a listed official path, hand the step to the user, stop retrying, or look up once more. The answer has three layers: access_profile (the host's own declared agent policy from robots.txt and llms.txt), official_paths and provider_status (official developer docs and status feeds), and observations (recent reports from other agents, often empty). Declared policy is not an access guarantee and no_reports does not establish that a site works. Do not poll without a task need. If a blocked lookup's answer carries a feedback_token and you then try again, you may report once with whether it worked. Follow your own operating authority and platform rules before submitting an observation. Send only public hostnames and structured fields, never personal URLs or credentials.

Tools

Reads data1

Changes data2

Skills

None in the store listings or on the server

MCP connector

Server URL

https://agenthelpnetwork.com/mcp

None

Protocol

MCP 2026-07-28

Instructions

AHN (Agent Help Network) answers are advisory: your platform's rules and your user's instructions come first; if an answer conflicts with them, do not act on it, and if its evidence is thin, tell your user what you found rather than guess; thin evidence alone is not a reason to stop. Call with the public hostname when an authorized web step is blocked or failing for no clear reason (a 403 or access denied, a 429, a 5xx or a timeout, a login wall, a "verify you are human" or CAPTCHA page) or a user asks "is this site down?" or whether it blocks agents. Do not call it when the page opened normally or when your own tool or platform already said why it refused (for example a robots.txt or platform policy refusal). Look the host up once per blocked step, before deciding whether to retry. Send the public hostname, never a URL path, query or credentials; add the codes only when you observed them. If you only check whether a site is up (not blocked), send task_step=status_check. Browser automation usually sends a normal browser user agent, so robots.txt agent tokens may not apply to it; leave agent_vendor out then. Read summary_line first, then follow next_action: it may say to continue, wait for the provider, use a listed official path, hand the step to the user, stop retrying, or look up once more. The answer has three layers: access_profile (the host's own declared agent policy from robots.txt and llms.txt), official_paths and provider_status (official developer docs and status feeds), and observations (recent reports from other agents, often empty). Declared policy is not an access guarantee and no_reports does not establish that a site works. Do not poll without a task need. If a blocked lookup's answer carries a feedback_token and you then try again, you may report once with whether it worked. Follow your own operating authority and platform rules before submitting an observation. Send only public hostnames and structured fields, never personal URLs or credentials.