// A robust methodology for LLM-based requirements gathering and high-fidelity artifact generation. Employs 'Divide, Conquer, and Verify' tactics using specialized subagents, iterative exhaustion loops, and batched execution to ensure zero data loss.
End-to-end GECX/CXAS/CES conversational agent lifecycle -- build agents from requirements (PRD-to-agent), create and run evals (goldens, simulations, tool tests, callback tests), debug failures, and iterate to production quality. Use this skill whenever the user mentions GECX, CXAS, CES, SCRAPI, conversational agents, voice agents, audio agents, agent evals, pushing/pulling/linting agents, or agent instructions/callbacks/tools on the Google Customer Engagement Suite platform.
Migrate Dialogflow CX (DFCX) agents to CXAS (Customer Experience Agent Studio) agents. Use this skill when the user mentions DFCX migration, migrating agents, converting DFCX to CXAS, porting agents, agent migration, or post-migration optimization/consolidation. Four independently runnable scripts: migrate.py (1:1), stage_1.py (variable dedup + consolidation), stage_2.py (instruction state machines + tool mocks + lint + report), stage_3.py (rewires consolidated topology from source dep graph; only needed when stage_1 ran consolidation). State persists between scripts via <target>_ir.json so each can run / re-run / resume independently.
Retrieves non-contained CCAI Insights conversations (losses), uses agent intelligence to cluster them into common failure patterns, and generates a markdown report. Use when you need to analyze failure patterns and build targeted regression/evaluation reports.
Automates the ingestion of customer requirement documents such as diagrams, BRDs, code etc., synthesizes high-fidelity natural transcripts, and compiles them into highly interactive, responsive Critical User Journey (CUJ) reports.
A two-phase protocol for extracting structure and generating transcripts from customer artifacts.
Enforces task coverage and prevents drift via a deterministic checklist tool with output verification.
| name | cxas-protocol-robust-extraction |
| description | A robust methodology for LLM-based requirements gathering and high-fidelity artifact generation. Employs 'Divide, Conquer, and Verify' tactics using specialized subagents, iterative exhaustion loops, and batched execution to ensure zero data loss. |
This protocol defines the standard operating procedure for extracting exhaustive requirements (like subintents, CUJs, or logic rules) from large, complex, or fragmented customer artifacts.
It prevents the common LLM pitfalls of "context drift" and "truncation" by enforcing a strict "Divide, Conquer, and Verify" methodology.
To ensure 100% coverage and prevent data loss due to tool limits or implicit filtering, follow these principles across all phases:
When tasked with comprehensive extraction or generation from a large corpus, you MUST follow this four-phase methodology:
Never use a single generalist agent or a single prompt to read all files.
cxas-ingestor-adk) in
parallel, providing each with only the context relevant to their expertise.LLMs often miss items in a single pass of a large document. You must force them to iterate.
Once the exhaustive list is finalized (e.g., 100+ subintents), organize it.
Never ask an LLM to generate 100+ complex artifacts (like conversational transcripts) in a single prompt. It will hallucinate or truncate.