Codex
Connect Codex to heyeveryone.
Bring content work from Codex into a workspace with a visible review and publishing flow.
Connect with the shared guide
Follow the connection steps below, then choose the workspaces the agent may access. These steps are imported from the maintained agent guide.
Try a draft before a schedule
Example prompt: “Draft a post from these notes for the channels connected to my workspace. Show me the content link and any channel checks. Do not schedule or publish.” Review the result in heyeveryone, then ask Codex to request a review if needed.
What stays with a person
An agent cannot sign off, create review links, manage tokens or change billing. Schedule requests wait in Updates for confirmation by default. A person confirms publishing by default. A workspace owner can explicitly enable auto-publish for a particular agent in a particular workspace.
Credits and access
Agent access is included in every plan. MCP and API access does not cost credits; actions that run heyeveryone’s own AI do. See the credit cost before you run them. Your agent provider may charge separately.
Connect Codex
OAuth:
codex mcp add heyeveryone --url https://heyeveryone.ai/mcp
codex mcp login heyeveryonePersonal access token:
export HEYEVERYONE_TOKEN=hey_pat_...
codex mcp add heyeveryone --url https://heyeveryone.ai/mcp \
--bearer-token-env-var HEYEVERYONE_TOKENor in ~/.codex/config.toml (or a project's .codex/config.toml):
[mcp_servers.heyeveryone]
url = "https://heyeveryone.ai/mcp"
bearer_token_env_var = "HEYEVERYONE_TOKEN"- In the TUI,
/mcplists connected servers. - IDE extension: gear menu > MCP servers > Add server > Streamable HTTP.
- The old
experimental_use_rmcp_clientflag is no longer needed for HTTP servers. - Codex speaks the
initializehandshake (MCP 2025-11-25) unless its 2026-07-28 feature flag is on. heyeveryone serves both.
Sources: learn.chatgpt.com: MCP in Codex, Codex config reference.