Decompose a raw prompt into structured blocks (role, objective, context, constraints, etc.). Uses AI (Claude/OpenAI) if an API key is configured on the server, otherwise falls back to keyword-based heuristic analysis. Returns a JSON list of blocks ready to edit or pass to compile_prompt. Args: pr...
AI agents call decompose_prompt to retrieve information from Flompt without modifying anything — typically the context-gathering step in research, monitoring, and reporting workflows, before the agent takes action elsewhere.
This tool takes a prompt string as input and analyzes/parses it into semantic blocks. It only reads and processes the provided text, returning structured data. There are no side effects, no data is stored or modified, and no external operations are triggered beyond the AI API call used for analysis.
From the tool's definition Decompose a raw prompt into structured blocks... Returns a JSON list of blocks ready to edit or pass to compile_prompt.
Documented attack patterns abuse exactly the kind of access decompose_prompt gives an agent:
PolicyLayer is an MCP gateway — it sits between your AI agents and Flompt, and nothing reaches the server without passing your rules. This is the rule we recommend for decompose_prompt:
{
"version": "1",
"default": "deny",
"tools": {
"decompose_prompt": {}
}
}decompose_prompt is read-only, so it stays allowed — but everything else on the server is denied unless you say otherwise.
Free to start. No card required.
Decompose a raw prompt into structured blocks (role, objective, context, constraints, etc.). Uses AI (Claude/OpenAI) if an API key is configured on the server, otherwise falls back to keyword-based heuristic analysis. Returns a JSON list of blocks ready to edit or pass to compile_prompt. Args: prompt: The raw prompt string to decompose. Returns: A summary of extracted blocks + the full JSON to pass to compile_prompt. It is categorised as a Read tool in the Flompt MCP Server, which means it retrieves data without modifying state.
Register the Flompt MCP server in PolicyLayer and add a rule for decompose_prompt: 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 Flompt. Nothing to install.
decompose_prompt is a Read tool with low risk. Read-only tools are generally safe to allow by default.
Yes. Add a rate_limit block to the decompose_prompt 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 decompose_prompt. 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.
decompose_prompt is provided by the Flompt MCP server (nyrok/flompt). PolicyLayer sits as a proxy in front of this server to enforce policies before tool calls reach the server.
Deterministic rules across all 3 Flompt tools. Per-identity grants. Full audit log. Live in minutes. Nothing to install.
Free to start. No card required.
3 Flompt tools catalogued and risk-classified — across an index of 42,500+ MCP servers.