Skip to main content
Glama

Server Configuration

Describes the environment variables required to run the server.

NameRequiredDescriptionDefault
dataNoThe data directory for persistence (default is ./data).
portNoThe port to run the server on (default is 3001).
boardNoSpecify the name or path for the shared blackboard.
ceilingNoSet the hard token ceiling for the federated budget.
no-tokenNoDisable HMAC-signed permission tokens.
audit-logNoSpecify the path for the append-only audit log file.
no-budgetNoDisable the federated token tracking/budget system.
no-controlNoDisable live orchestrator configuration tools.
OPENAI_API_KEYNoAPI key required if running agents or pipelines that utilize OpenAI models.

Capabilities

Features and capabilities supported by this server

CapabilityDetails
tools
{}

Tools

Functions exposed to the LLM to take actions

NameDescription
blackboard_read

Read a single entry from the shared blackboard by key. Returns the value, source agent, and timestamp, or null if not found.

blackboard_write

Write a value to the shared blackboard. Requires agent identity verification. Returns the written entry.

blackboard_list

List all keys currently on the blackboard. Optionally filter by key prefix.

blackboard_delete

Delete an entry from the blackboard by key. Agent must have write access to the key.

blackboard_exists

Check whether a key exists on the blackboard (and has not expired).

budget_status

Get the current budget status: ceiling, total spent, remaining tokens, and per-agent breakdown.

budget_spend

Spend tokens on behalf of an agent. Returns allowed/denied with remaining balance.

budget_reset

Reset all spend counters to zero. The ceiling is preserved. Use when starting a new task cycle.

budget_set_ceiling

Dynamically change the global token ceiling. Can be raised or lowered at runtime.

budget_get_log

Retrieve the full transaction log of all spend() calls in chronological order.

token_create

Generate a new signed security token for an agent. Returns the full token object.

token_validate

Validate a security token — checks signature, expiration, and revocation status.

token_revoke

Revoke a token by its tokenId. The token will be rejected on all future validation attempts.

audit_query

Query the audit log. Filter by agent_id, event_type, outcome, or time range.

audit_tail

Return the N most recent audit log entries. Fast — reads from end of file.

config_get

Read the current Network-AI orchestrator configuration (maxParallelAgents, defaultTimeout, enableTracing, etc.).

config_set

Update a Network-AI configuration value at runtime. Changes take effect immediately for all subsequent operations.

agent_list

List all agents currently registered with the orchestrator, including their status and last activity.

agent_spawn

Dispatch a task to a named agent by writing a task record to the blackboard. The agent will pick it up on its next poll cycle.

agent_stop

Signal an agent to stop by writing a stop record to the blackboard and updating its status.

fsm_transition

Manually advance an FSM (Finite State Machine) to a new state by writing the transition to the blackboard.

orchestrator_info

Get a snapshot of the entire orchestrator: version, config, agent count, blackboard key count, and system health.

Prompts

Interactive templates invoked by user choice

NameDescription

No prompts

Resources

Contextual data attached and managed by the client

NameDescription

No resources

Latest Blog Posts

MCP directory API

We provide all the information about MCP servers via our MCP API.

curl -X GET 'https://glama.ai/api/mcp/v1/servers/Jovancoding/network-ai'

If you have feedback or need assistance with the MCP directory API, please join our Discord server