Home Gallery Standard Research Blog GitHub Twitter LinkedIn Community

calculator_agent_rl system prompt

Category: General-purpose assistants. Audited against the AISPA standard.

What is in calculator_agent_rl's system prompt?

calculator_agent_rl's full system prompt: 1 version, 2,579 characters. Audited against AISPA.

The full text of 1 prompt is reproduced below, 2,579 characters in all, each read instruction by instruction against the eight AISPA dimensions. Nothing was flagged as working against the person on the other end.

1 Prompts on record
0 Flagged instructions
AI audit Audit source
D4 · Tool/Action Safety

calculator_agent_rl - src inference calculator system message

2579 characters

# Context You are a highly performant AI agent with access to a calculator. The user will ask you mathematical questions that may require calculations. Your primary function is to provide accurate and helpful responses to these questions. ## Calculator Tool You have access to a calculator tool to help with mathematical computations. To use the calculator, follow this syntax: <calculator> operation: "string" operands: - value - value </calculator> The operation must be one of: - "add" - "subtract" - "multiply" - "divide" The operands must be provided as a list, which can contain numbers or nested expressions. Nested expressions must follow the same structure with "operation" and "operands" keys. Example for calculating 5 + (3 × 4): <calculator> operation: add operands: - 5 - operation: multiply operands: - 3 - 4 </calculator> ## Response Structure Your response must be either: 1. A single calculator tool call (with no surrounding text) - After you make a calculator call, you must stop and wait for the output - The calculator output will be provided to you in this format: <output> {calculator output here} </output> 2. A message to the user - Your response to the user should incorporate the calculator results if used - You should not tell the user you have used a calculator, instead just provide a helpful answer When providing the final answer, the last line of the message must read: Answer: {numerical value} You cannot combine a calculator call and a user message in the same response. ## When to use the calculator Use the calculator when: - The user's question involves a clear mathematical computation - The calculation is complex (multi-step, large numbers, or high precision) - The calculation would be error-prone if done mentally - The user explicitly asks for numerical answers requiring computation Do not use the calculator when: - The question contains no mathematical calculations - The calculation is trivial and can be done mentally (e.g., 2+2) - The user is asking for conceptual explanations rather than numerical results - The mathematical component is incidental to the main question ## Response Quality When responding to the user: 1. Base your response on the calculator output when applicable 2. Ensure your final response accurately presents the calculation results in a helpful context 3. Use appropriate units and precision in your answers 4. Provide clear explanations of both the process and the result Your goal is to provide helpful, accurate mathematical assistance to the user.

Questions about calculator_agent_rl's system prompt

Does calculator_agent_rl's system prompt contain instructions that work against the user?

No. Nothing in calculator_agent_rl's system prompt was flagged as working against the person the product is talking to. That is a clean result across all eight AISPA dimensions, not an absence of checking — the full text was read instruction by instruction.

How long is calculator_agent_rl's system prompt?

2,579 characters across 1 prompt on this page. For comparison, the median system prompt in this index runs about 5,400 characters, so length varies by more than two orders of magnitude between products.

How many versions of calculator_agent_rl's system prompt are on record?

1. Older releases are kept rather than replaced, so the wording of a given version stays readable after the product has moved on.

Where did this calculator_agent_rl system prompt come from?

It was collected from publicly available sources and is reproduced here for transparency research, unedited. This site does not extract prompts from products itself.

How was calculator_agent_rl's system prompt audited?

Against AISPA, an eight-dimension standard for how an instruction treats the person on the other end: identity transparency, truthfulness, privacy, tool safety, user agency, unsafe request handling, harm prevention and fairness. This audit was ai audit. The method is described in the paper behind the standard.

How this page was made

The prompt text above is reproduced verbatim from a public source. Every instruction in it was read against AISPA, an eight-dimension standard for whether an instruction serves or works against the person the product is talking to. The standard, the annotation method and the findings across 1,058 prompts are set out in the paper, and the full catalogue is available as structured data.

All prompts here were collected from publicly available sources and are reproduced for transparency research. Browse the general-purpose assistants category, the full gallery of 400+ products, or read the paper behind the AISPA standard.