AI agents use update_sift to create or update resources in Sifter — usually the action step of a workflow, after the agent has gathered context. Every call changes real data in your Sifter environment.
This tool modifies metadata (name and instructions) of an existing sift record without deleting or destroying data. The changes are reversible—a sift's name or instructions can be changed again. This is a classic Write operation.
From the tool's definition Tool name is 'update_sift' and description states it 'Update[s] an existing sift's name or instructions.' The word 'update' indicates modification of existing data in a reversible manner.
Documented attack patterns abuse exactly the kind of access update_sift gives an agent:
PolicyLayer is an MCP gateway — it sits between your AI agents and Sifter, and nothing reaches the server without passing your rules. This is the rule we recommend for update_sift:
{
"version": "1",
"default": "deny",
"tools": {
"update_sift": {
"limits": [
{
"counter": "update_sift_rate",
"window": "minute",
"max": 30,
"scope": "grant"
}
]
}
}
}update_sift 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.
Update an existing sift's name or instructions. It is categorised as a Write tool in the Sifter MCP Server, which means it can create or modify data. Consider rate limits to prevent runaway writes.
Register the Sifter MCP server in PolicyLayer and add a rule for update_sift: 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 Sifter. Nothing to install.
update_sift 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 update_sift 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 update_sift. 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.
update_sift is provided by the Sifter MCP server (sifter-ai/sifter). PolicyLayer sits as a proxy in front of this server to enforce policies before tool calls reach the server.
Deterministic rules across all 15 Sifter tools. Per-identity grants. Full audit log. Live in minutes. Nothing to install.
Free to start. No card required.
15 Sifter tools catalogued and risk-classified — across an index of 42,500+ MCP servers.