Reads or updates scoring weight configuration for get_relevant_context. Omit weights to read current config; provide a JSON string of weight overrides to update. Valid factors: semantic_similarity, recency, reference_frequency, lifecycle_status, scope_priority, code_proximity, explicit_priority. ...
AI agents use manage_budget_config to create or update resources in Engrams — usually the action step of a workflow, after the agent has gathered context. Every call changes real data in your Engrams environment.
This tool has dual Read/Write capability, but the presence of configuration updates ('updates scoring weight configuration', 'provide...weight overrides to update') classifies it as Write rather than Read. Modifying scoring weights for context retrieval could affect which information an AI agent retrieves and prioritizes, potentially causing the system to surface irrelevant or manipulated context.
From the tool's definition Tool description states it 'Reads or updates scoring weight configuration' and 'provide a JSON string of weight overrides to update.' The update capability makes this a Write operation.
Documented attack patterns abuse exactly the kind of access manage_budget_config gives an agent:
PolicyLayer is an MCP gateway — it sits between your AI agents and Engrams, and nothing reaches the server without passing your rules. This is the rule we recommend for manage_budget_config:
{
"version": "1",
"default": "deny",
"tools": {
"manage_budget_config": {
"limits": [
{
"counter": "manage_budget_config_rate",
"window": "minute",
"max": 30,
"scope": "grant"
}
]
}
}
}manage_budget_config stays usable, but capped — an agent stuck in a loop can't make hundreds of changes a minute. Everything else on the server is denied unless you say otherwise.
Free to start. No card required.
Reads or updates scoring weight configuration for get_relevant_context. Omit weights to read current config; provide a JSON string of weight overrides to update. Valid factors: semantic_similarity, recency, reference_frequency, lifecycle_status, scope_priority, code_proximity, explicit_priority. Each 0.0-1.0. Returns: {weights: {...}, source:. It is categorised as a Write tool in the Engrams MCP Server, which means it can create or modify data. Consider rate limits to prevent runaway writes.
Register the Engrams MCP server in PolicyLayer and add a rule for manage_budget_config: allow, deny, rate-limit, or require approval. Point your MCP client at the PolicyLayer proxy URL and the rule is enforced on every call, before it reaches Engrams. Nothing to install.
manage_budget_config is a Write tool with medium risk. Write tools should be rate-limited to prevent accidental bulk modifications.
Yes. Add a rate_limit block to the manage_budget_config rule in your PolicyLayer policy. For example, setting max: 10 and window: 60 limits the tool to 10 calls per minute. Rate limits are tracked per agent session and reset automatically.
Set action: deny in the PolicyLayer policy for manage_budget_config. The AI agent will receive a policy violation error and cannot call the tool. You can also include a reason field to explain why the tool is blocked.
manage_budget_config is provided by the Engrams MCP server (stevebrownlee/engrams). PolicyLayer sits as a proxy in front of this server to enforce policies before tool calls reach the server.
Start from Engrams, add the rest of your stack, and see everything your agents can call. Then put policy on all of it.
Free to start. No card required.
42 Engrams tools catalogued and risk-classified — across an index of 43,000+ MCP servers.