UNPKG

@blundergoat/goat-flow

Version:

AI coding agent harness and local dashboard for Claude Code, OpenAI Codex, Google Antigravity, and GitHub Copilot - setup audits, guardrails, structured skills, deny hooks, and persistent learning loops.

9 lines 458 B
/** * Context concern: Is the agent's map accurate and structurally complete? * 5 deterministic checks (instruction size, execution loop, doc paths, * instruction sections, boundary guidance). Content-quality judgments (e.g. footgun evidence * currency) live in the `quality` assessment prompt, not here. */ import type { HarnessCheck } from "../types.js"; export declare const CONTEXT_CHECKS: HarnessCheck[]; //# sourceMappingURL=check-context.d.ts.map