Identity & Introduction
src/constants/prompts.ts:175
The identity section is deliberately minimal — a single sentence that establishes role (interactive agent), domain (software engineering), and deference to the rest of the prompt ('use the instructions below'). This avoids over-constraining the persona early, which research shows can cause the model to ignore later instructions. The CYBER_RISK_INSTRUCTION is inlined immediately to establish security boundaries before any task-specific guidance. The URL guardrail prevents a common failure mode where models hallucinate plausible-looking links.
Techniques Used
Tags
Appears in use cases
This prompt is a step in curated flows that show how pieces of Claude Code connect for real tasks.
Fixing a Bug
From symptom to patch using search, read, edit, and verification
Exploring a Codebase
Discovery with glob, grep, reads, and optional Explore agent
Planning & Task Management
Plan mode, Plan agent, todos, and background tasks
Multi-Agent Orchestration
Coordinator mode, Agent tool, and worker personas
Memory & Context
Environment injection, memory taxonomy, filtering, extraction, and output shaping
Web Research & Integration
Search, fetch, then write and validate in the repo