Use when a runner agent receives a distance update and must deplete water proportionally, or when the runner enters a hydration station and must decide whether to stop and rehydrate. Triggered by distance events and station events.
Use when a runner agent needs to manage athletic performance during a race tick: accelerating, braking, or reading current vitals (speed, energy, hydration, exhaustion). Triggered every tick the runner is active.
Use when the simulator's loop agent advances the race by one tick: broadcasting state to runners, draining the telemetry buffer, computing traffic conditions, and checking for race completion. Triggered every iteration of the LoopAgent.
Use when the simulator finishes its race-tick loop and must wrap up: compiling final results from tick snapshots, stopping the RaceCollector, and reporting outcomes to other agents. Triggered once per simulation, after the last tick.
Use when the simulator receives a validated plan from the planner and must initialize the race: parsing the plan, spawning runner agents, starting the telemetry collector, and firing the start gun. Triggered once per simulation, before the first tick.
Use when running the simulator-with-failure test variant and the pre-race phase begins. The preparesimulation tool intentionally raises RuntimeError to exercise the ADK and SimulationCommunicationPlugin toolerror callback path. Not intended for production simulations.
Use when an agent renders rich UI back to a client surface (cards, dashboards, forms, modals) using the A2UI v0.8.0 declarative protocol. Required whenever the response should display structured components rather than plain text, or whenever the validateandemita2ui tool is in scope.
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…
Calculates and generates static coverage reports for Gemini Enterprise for Customer Experience (GECX) conversational agents by mapping evaluations against tools, callbacks, agent transfers and instructions. Use when analyzing unit test / evaluation coverage of a conversational agent or identifying gap areas (un-tested…
Generate eval YAMLs for one entire eval type (all goldens, all sims, all tooltests, or all callbacktests) in a single dispatch. Reads the TDD's Coverage Map, the agent's actual tools and variables, then writes the appropriate file(s) — see "File layout per type" for what each type requires (sims are one file by runner…
Run cxas lint (with optional scoping to a set of agents or tools), fix every error using the rule recipes in references/build.md, verify each Edit landed on disk, and re-lint until clean. Use after a fresh build or major edit when the lint output has multiple errors and the fixes are mechanical.
Produce a Technical Design Document (TDD) for a GECX agent. Auto-detects mode from inputs — reverse-engineer from an existing app's code, OR draft from scratch given any combination of requirements artifacts (PRD, spec, sample conversations, mock data, customer profiles, reference agents). Returns the TDD content plus…
Diagnose ONE failing GECX eval. Read its transcript, classify against the decision tree, and return a structured diagnosis JSON the main thread can act on. Fan out in parallel for the top N failures from a triage run.
Author, validate, and manage Contact Center AI (CCAI) Insights Autolabeling Rules. Use when users want to translate business labeling logic into CEL condition expressions, maintain declarative autolabelrules.yaml configurations, or synchronize rules to GCP projects.
Author, validate, and manage Contact Center AI (CCAI) Insights Configurable Dashboards. Use when users want to define multi-tab analytics dashboards, configure Vega-Lite charts and SQL queries, maintain declarative dashboards.yaml configurations, or synchronize dashboards to GCP projects.
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.
during Phase 2, extracts and normalizes the Critical User Journey (CUJ) taxonomy fields, deduplicates overlapping categories, and grounds them in the high-level domain context.
Analyzes a specific set of artifacts (e.g., Cyara test cases, Drawio diagrams, ADK code, DFCX declarative page packages) to recursively extract granular sub-intents and dialogue flows.
Perform an expert manual and tool-based architectural inspection of the target requirements directory (using your standard listdir, codesearch, or find tools to crawl the folder tree).