Research Agent Brief: Source Discovery, Evidence Review, and Citation Table
Directs an AI agent to plan source discovery, grade evidence, build a citation table, and support a decision without claiming live browsing occurred.
217 prompts
Prompts written for Claude, Anthropic's assistant. Claude handles long inputs well, so many of these prompts ask you to paste a whole document, transcript or code excerpt and then structure the answer around it.
8 results
Directs an AI agent to plan source discovery, grade evidence, build a citation table, and support a decision without claiming live browsing occurred.
Produces a least-privilege tool inventory, function definitions, and approval rules for an agent that calls APIs or internal systems.
Designs verification checks, error handling, and a self-correction loop for an agent task before it runs, based on likely failure modes.
Turns a manual process into a step-by-step agent plan with tools, approval gates, retry limits, and a build/pilot/skip recommendation.
Turns an agent idea into a build-ready design: the job, the tools with their inputs and permissions, guardrails, failure handling and a set of evaluation cases to test it.
Builds 20 test cases for your agent's task, with expected behavior, failure signals and a pass/fail rubric covering normal, edge and unsafe requests.
Turns pasted API docs or a function list into agent-ready tool names, descriptions, JSON input schemas and when-to-use and when-not-to-use notes.
Produces a set of testable guardrail rules for your coding agent, covering blocked commands, approval gates, secret handling and hook checks.
Put the documents first and the prompt after them. Anthropic's prompting guide suggests wrapping each document in tags such as <document> so Claude keeps the sources apart.
Yes. Paste it into the project instructions and every chat in that project starts from it. Fill its blanks with the details that stay the same from chat to chat.
No. These are for the Claude apps and the API. For work inside a code repository, where the agent reads files and runs commands, use the Claude Code prompts.