routing-policy-change
$
npx mdskill add vllm-project/semantic-router/routing-policy-change- Change boolean decision logic after signal extraction - Change candidate-model selection, ranking, cost routing, or latency routing - Change looper behavior that belongs to matched-decision policy instead of signal extraction or plugin hooks
SKILL.md
.github/skills/routing-policy-changeView on GitHub ↗
--- name: routing-policy-change category: primary description: Modifies routing policy after signal extraction, including matched-decision logic, candidate-model selection, and downstream looper behavior. Use when changing decision predicates, thresholds, priorities, model ranking, cost or latency routing, or other post-signal routing policy. --- # Routing Policy Change ## Trigger - Change boolean decision logic after signal extraction - Change candidate-model selection, ranking, cost routing, or latency routing - Change looper behavior that belongs to matched-decision policy instead of signal extraction or plugin hooks ## Workflow 1. Read change surfaces and module boundaries for routing-policy context 2. Modify decision predicates, downstream model selection, or matched-decision looper behavior 3. Run `make agent-report ENV=cpu CHANGED_FILES="..."` to identify impacted surfaces 4. Run `make agent-ci-gate CHANGED_FILES="..."` to validate all affected constraints 5. Verify routing-policy behavior is covered by targeted tests and affected E2E expectations ## Gotchas - Keep signal extraction upstream; do not smuggle classifier changes into this primary unless the signal contract itself is changing. - Keep plugin hooks separate; looper changes that only support matched-decision policy belong here, but request or response body plugins still belong to the plugin flow. - Decision predicates and candidate-model ranking can change user-visible routing even when the patch looks small, so they are not refactors by default. ## Must Read - [docs/agent/change-surfaces.md](../../../../docs/agent/change-surfaces.md) - [docs/agent/module-boundaries.md](../../../../docs/agent/module-boundaries.md) - [docs/agent/testing-strategy.md](../../../../docs/agent/testing-strategy.md) ## Standard Commands - `make agent-report ENV=cpu CHANGED_FILES="..."` - `make agent-ci-gate CHANGED_FILES="..."` - `make agent-feature-gate ENV=cpu CHANGED_FILES="..."` ## Acceptance - Decision predicates and downstream candidate-model selection stay aligned with the same routing-policy intent - Targeted tests and affected E2E coverage are updated when routing-policy behavior changes - Routing-policy logic stays downstream of signal extraction and separate from plugin-only processing
More from vllm-project/semantic-router
- config-platform-changeSynchronizes config representations across router config, Python CLI schema, and dashboard config UI. Use when adding or changing a config concept that spans those surfaces or addressing config representation debt before Kubernetes-facing translation.
- cross-stack-bugfixDiagnoses and fixes bugs that span multiple layers (runtime, CLI, UI, platform, tests) requiring coordinated changes across surfaces. Use when a bug does not map cleanly to a narrower skill, the fix touches more than one surface, or changes need cross-cutting validation.
- dashboard-platform-changeModifies dashboard frontend or backend surfaces that present, configure, or manage router behavior through the console UI. Use when changing dashboard pages or components, backend handlers, console persistence, auth or session flows, or user-visible routing metadata in the dashboard.
- fleet-sim-changeModifies the fleet simulator package, API service, release wiring, or simulator-owned docs and assets as one maintained subsystem. Use when changing src/fleet-sim, simulator release workflow, or fleet-sim-owned docs and assets under website/.
- harness-contract-changeModifies the repository's agent contract including AGENTS.md, docs index, manifests, validation scripts, and contributor-facing harness wrappers. Use when updating agent documentation, changing repo manifests, editing validation scripts, modifying CI/workflow classification, or updating contributor-facing guides like README.md, CONTRIBUTING.md, or the PR template.
- k8s-platform-changeModifies Kubernetes-facing operator, CRD, deployment-profile, or DSL translation behavior for semantic-router platform integration. Use when changing operator APIs or controllers, deployment stack manifests, profile-owned platform wiring, or router-to-Kubernetes translation layers.
- maintainer-issue-pr-managementManages GitHub issue and pull-request lifecycle including creation, updates, triage labelling, and closeout metadata using canonical templates and repository taxonomy. Use when a maintainer asks to create, update, close, or triage GitHub issues or PRs, or when issue creation requires codebase analysis for scope, labels, or acceptance criteria.
- maintainer-release-opsMaintainer release and milestone operating workflow. Use when a maintainer wants to plan a release, create milestone issues, sync GitHub issue or PR state, generate a daily review brief, or manage stale PRs and backlog routing.
- openclaw-vsr-bridgeInstall vLLM Semantic Router in agent-safe mode, import supported OpenClaw model providers into canonical VSR config, and rewrite OpenClaw to target VSR.
- plugin-end-to-endImplements end-to-end plugin changes spanning router config, post-decision processing, optional CLI/UI exposure, and E2E test coverage. Use when adding a new plugin type, changing plugin config schema or execution semantics, updating plugin chain behavior, or modifying plugin-exposed metadata across surfaces.