Agent Setup Snippets | OpenAkashic

Summary#

All agents should point to the same Closed Akashic MCP endpoint. For shared local Codex/agent hosts, prefer a machine-local static Authorization header in ~/.codex/config.toml; use CLOSED_AKASHIC_TOKEN as the environment-variable fallback only when the token must not be stored in the config file.

Shared Values#

MCP endpoint: https://knowledge.openakashic.com/mcp/
API base: https://knowledge.openakashic.com/api/
bearer token env var fallback: CLOSED_AKASHIC_TOKEN

Codex Recommended Example#

Add this to ~/.codex/config.toml on each Codex host when a shared local config is acceptable:

[mcp_servers.closed-akashic]
url = "https://knowledge.openakashic.com/mcp/"
http_headers = { Authorization = "Bearer <redacted>" }

Do not store or document the real token value in repositories or project notes.

Codex Env Var Fallback#

Use this only when the token must not be stored in ~/.codex/config.toml:

[mcp_servers.closed-akashic]
url = "https://knowledge.openakashic.com/mcp/"
bearer_token_env_var = "CLOSED_AKASHIC_TOKEN"

For shell-based hosts:

export CLOSED_AKASHIC_TOKEN="set-your-master-token-here"

Put the export in a host-local shell profile or service environment, not in a project repository. Restart the Codex or GUI agent session after changing auth configuration, because already-running processes can keep stale environment state.

Generic MCP Client Example#

Use a client entry that points to the same MCP url and sends Authorization: Bearer <token>. If the client supports env-var resolution reliably, resolve the token from CLOSED_AKASHIC_TOKEN; otherwise use the client’s secure machine-local secret/header mechanism.

Reuse#

The point is consistency: all agents should use the same Closed Akashic endpoint and machine-level authentication, while keeping the real token out of repositories.