https://praxis.fast/ai/dev/examples/configs/Recent content in All configurations on PraxisHugoenhttps://praxis.fast/ai/dev/examples/configs/a2a-agent-card-routing.yaml/Mon, 01 Jan 0001 00:00:00 +0000https://praxis.fast/ai/dev/examples/configs/a2a-agent-card-routing.yaml/Routes agent card discovery requests to dedicated backendshttps://praxis.fast/ai/dev/examples/configs/a2a-classifier-routing.yaml/Mon, 01 Jan 0001 00:00:00 +0000https://praxis.fast/ai/dev/examples/configs/a2a-classifier-routing.yaml/Routes A2A requests by body-derived method, family, context ID, task ID, and streaming detectionhttps://praxis.fast/ai/dev/examples/configs/a2a-task-routing.yaml/Mon, 01 Jan 0001 00:00:00 +0000https://praxis.fast/ai/dev/examples/configs/a2a-task-routing.yaml/Captures task and context ownership from SendMessage JSON responses and SendStreamingMessage / SubscribeToTask SSE responses, then routes follow-up requests back to the backend cluster that created the task or owns the contexthttps://praxis.fast/ai/dev/examples/configs/ai-inference-body-based-routing.yaml/Mon, 01 Jan 0001 00:00:00 +0000https://praxis.fast/ai/dev/examples/configs/ai-inference-body-based-routing.yaml/Routes LLM API requests to different backends based on the model field in the JSON request bodyhttps://praxis.fast/ai/dev/examples/configs/aws-sigv4.yaml/Mon, 01 Jan 0001 00:00:00 +0000https://praxis.fast/ai/dev/examples/configs/aws-sigv4.yaml/Signs outbound requests to an AWS service (Bedrock, in this example) using Signature Version 4. Credentials are static, sourced from environment variables — see the module docs on Sigv4SignFilter for the planned OIDC/default-credential-chain follow-uphttps://praxis.fast/ai/dev/examples/configs/azure-ad.yaml/Mon, 01 Jan 0001 00:00:00 +0000https://praxis.fast/ai/dev/examples/configs/azure-ad.yaml/Acquires an Entra ID bearer token via the client-credentials grant and injects “Authorization: Bearer " on every proxied request to Azure OpenAIhttps://praxis.fast/ai/dev/examples/configs/credential-injection.yaml/Mon, 01 Jan 0001 00:00:00 +0000https://praxis.fast/ai/dev/examples/configs/credential-injection.yaml/Injects per-cluster API credentials into upstream requests and strips client-provided credentials to prevent forwardinghttps://praxis.fast/ai/dev/examples/configs/external-metering.yaml/Mon, 01 Jan 0001 00:00:00 +0000https://praxis.fast/ai/dev/examples/configs/external-metering.yaml/Pre-request balance check and post-response token usage reporting against an external metering servicehttps://praxis.fast/ai/dev/examples/configs/external-metering-file-key.yaml/Mon, 01 Jan 0001 00:00:00 +0000https://praxis.fast/ai/dev/examples/configs/external-metering-file-key.yaml/Same pre-request balance check and post-response usage reporting as external-metering.yaml, but the internal bearer credential used to authenticate the metering callouts is read from a file rather than embedded in the config or imagehttps://praxis.fast/ai/dev/examples/configs/file-descriptor-limits.yaml/Mon, 01 Jan 0001 00:00:00 +0000https://praxis.fast/ai/dev/examples/configs/file-descriptor-limits.yaml/Size and protect the descriptor budget of a metered gateway: pin the process limit, shed requests with 503 before descriptors run out, cap concurrent requests and metering callouts, and close idle keep-alive clients and pooled upstream connections so they cannot pin descriptorshttps://praxis.fast/ai/dev/examples/configs/gcp-adc.yaml/Mon, 01 Jan 0001 00:00:00 +0000https://praxis.fast/ai/dev/examples/configs/gcp-adc.yaml/Acquires an OAuth2 access token from the GCE/GKE metadata server or a service-account key file and injects “Authorization: Bearer " on every proxied request to Vertex AIhttps://praxis.fast/ai/dev/examples/configs/identity-header-guard.yaml/Mon, 01 Jan 0001 00:00:00 +0000https://praxis.fast/ai/dev/examples/configs/identity-header-guard.yaml/Captures identity headers matching a prefix into filter metadata and strips them before forwarding upstreamhttps://praxis.fast/ai/dev/examples/configs/inflight-tracking.yaml/Mon, 01 Jan 0001 00:00:00 +0000https://praxis.fast/ai/dev/examples/configs/inflight-tracking.yaml/Tracks live request counts and summed token reservations per model in an in-memory InFlightRegistryhttps://praxis.fast/ai/dev/examples/configs/intelligent-route-all-capabilities.yaml/Mon, 01 Jan 0001 00:00:00 +0000https://praxis.fast/ai/dev/examples/configs/intelligent-route-all-capabilities.yaml/Demonstrates every candidate capability and selection input handled by intelligent_route todayhttps://praxis.fast/ai/dev/examples/configs/intelligent-route-inference.yaml/Mon, 01 Jan 0001 00:00:00 +0000https://praxis.fast/ai/dev/examples/configs/intelligent-route-inference.yaml/Use model identity to select the matching inference backend.https://praxis.fast/ai/dev/examples/configs/intelligent-route-management-skip.yaml/Mon, 01 Jan 0001 00:00:00 +0000https://praxis.fast/ai/dev/examples/configs/intelligent-route-management-skip.yaml/Demonstrates the intelligent_route management-path skip list: management and discovery endpoints (model listing, subscriptions, API-key management, health) bypass model resolution entirely, while inference paths are still resolved by model and fail closed on an unknown modelhttps://praxis.fast/ai/dev/examples/configs/intelligent-route-mcp.yaml/Mon, 01 Jan 0001 00:00:00 +0000https://praxis.fast/ai/dev/examples/configs/intelligent-route-mcp.yaml/Routes MCP tools/call requests to the cluster that owns the requested tool, using the mcp.name metadata set by the mcp filterhttps://praxis.fast/ai/dev/examples/configs/intelligent-route-overlay.yaml/Mon, 01 Jan 0001 00:00:00 +0000https://praxis.fast/ai/dev/examples/configs/intelligent-route-overlay.yaml/Routes requests using a routing overlay file (routing-overlay.json) instead of inline YAML candidates. The overlay is rendered by the operator into a Kubernetes ConfigMap and projected as a volume mounthttps://praxis.fast/ai/dev/examples/configs/intelligent-route-projected-credentials.yaml/Mon, 01 Jan 0001 00:00:00 +0000https://praxis.fast/ai/dev/examples/configs/intelligent-route-projected-credentials.yaml/Starts with an empty credential table and injects a provider credential that is resolved on demand from a projected Kubernetes Secret mounthttps://praxis.fast/ai/dev/examples/configs/json-rpc-routing.yaml/Mon, 01 Jan 0001 00:00:00 +0000https://praxis.fast/ai/dev/examples/configs/json-rpc-routing.yaml/Routes JSON-RPC 2.0 requests to different backends based on the “method” field in the JSON request bodyhttps://praxis.fast/ai/dev/examples/configs/lakera-guard.yaml/Mon, 01 Jan 0001 00:00:00 +0000https://praxis.fast/ai/dev/examples/configs/lakera-guard.yaml/Screens every request body through Lakera Guard for content moderation before forwarding to the upstreamhttps://praxis.fast/ai/dev/examples/configs/llmd-ext-proc-routing.yaml/Mon, 01 Jan 0001 00:00:00 +0000https://praxis.fast/ai/dev/examples/configs/llmd-ext-proc-routing.yaml/A real llm-d EPP or test processor returns the trusted x-gateway-destination-endpoint headerhttps://praxis.fast/ai/dev/examples/configs/llmisvc-model-provider-resolver.yaml/Mon, 01 Jan 0001 00:00:00 +0000https://praxis.fast/ai/dev/examples/configs/llmisvc-model-provider-resolver.yaml/Rewrites publisher-ID body model values to the short model name for LLMISvc / KServe routing; the routing header (default X-Model) is left unchanged so routing can still use the publisher IDhttps://praxis.fast/ai/dev/examples/configs/mcp-classifier-routing.yaml/Mon, 01 Jan 0001 00:00:00 +0000https://praxis.fast/ai/dev/examples/configs/mcp-classifier-routing.yaml/Routes MCP requests by body-derived method and tool namehttps://praxis.fast/ai/dev/examples/configs/mcp-stateless-broker.yaml/Mon, 01 Jan 0001 00:00:00 +0000https://praxis.fast/ai/dev/examples/configs/mcp-stateless-broker.yaml/Configurable stateless MCP broker using the final MCP 2026-07-28 stateless profilehttps://praxis.fast/ai/dev/examples/configs/model-to-header-routing.yaml/Mon, 01 Jan 0001 00:00:00 +0000https://praxis.fast/ai/dev/examples/configs/model-to-header-routing.yaml/Routes LLM API requests to different backends based on the “model” field in the JSON request bodyhttps://praxis.fast/ai/dev/examples/configs/model-to-header-trust-existing.yaml/Mon, 01 Jan 0001 00:00:00 +0000https://praxis.fast/ai/dev/examples/configs/model-to-header-trust-existing.yaml/Routes on a model header that a trusted hop in front of Praxis has already set, skipping the filter entirely when the header is presenthttps://praxis.fast/ai/dev/examples/configs/model-to-provider.yaml/Mon, 01 Jan 0001 00:00:00 +0000https://praxis.fast/ai/dev/examples/configs/model-to-provider.yaml/Map a stable public model ID to an existing upstream cluster and target modelhttps://praxis.fast/ai/dev/examples/configs/nemo-guardrails.yaml/Mon, 01 Jan 0001 00:00:00 +0000https://praxis.fast/ai/dev/examples/configs/nemo-guardrails.yaml/Evaluates incoming requests against a NeMo Guardrails servicehttps://praxis.fast/ai/dev/examples/configs/nemo-guardrails-response.yaml/Mon, 01 Jan 0001 00:00:00 +0000https://praxis.fast/ai/dev/examples/configs/nemo-guardrails-response.yaml/Evaluates upstream responses against a NeMo Guardrails servicehttps://praxis.fast/ai/dev/examples/configs/openai-selected-provider-legacy-fixture.yaml/Mon, 01 Jan 0001 00:00:00 +0000https://praxis.fast/ai/dev/examples/configs/openai-selected-provider-legacy-fixture.yaml/Controlled inference-fixture replay for OpenAI Chat Completions to an Anthropic Messages providerhttps://praxis.fast/ai/dev/examples/configs/openai-selected-provider-legacy-translation.yaml/Mon, 01 Jan 0001 00:00:00 +0000https://praxis.fast/ai/dev/examples/configs/openai-selected-provider-legacy-translation.yaml/OpenAI Chat Completions clients reach the overlay-selected Anthropic Messages or OpenAI-compatible provider and receive translated Anthropic JSON/SSE or unchanged OpenAI-compatible responses. intelligent_route binds the trusted provider profile to each request before response and SSE processinghttps://praxis.fast/ai/dev/examples/configs/project-state-owner-headers.yaml/Mon, 01 Jan 0001 00:00:00 +0000https://praxis.fast/ai/dev/examples/configs/project-state-owner-headers.yaml/Demonstrates the production boundary used when an external authenticator injects separate tenant and subject headers. state_owner consumes those assertions into an immutable internal owner and strips the inbound copies. project_state_owner_headers then recreates destination-specific headers from that normalized contexthttps://praxis.fast/ai/dev/examples/configs/prompt-enrichment.yaml/Mon, 01 Jan 0001 00:00:00 +0000https://praxis.fast/ai/dev/examples/configs/prompt-enrichment.yaml/Injects system messages into OpenAI-compatible chat completion requests before forwarding to the upstream providerhttps://praxis.fast/ai/dev/examples/configs/provider-route.yaml/Mon, 01 Jan 0001 00:00:00 +0000https://praxis.fast/ai/dev/examples/configs/provider-route.yaml/This listener requires downstream mTLS. peer_identity_trust authenticates and authorizes the edge gateway before AI-owned x-ai-routing- fields can influence provider-local routinghttps://praxis.fast/ai/dev/examples/configs/stream-usage-inject.yaml/Mon, 01 Jan 0001 00:00:00 +0000https://praxis.fast/ai/dev/examples/configs/stream-usage-inject.yaml/Ensures every streaming OpenAI Chat Completions request carries stream_options.include_usage = true so the upstream response includes token usage in the final SSE eventhttps://praxis.fast/ai/dev/examples/configs/time-to-first-token.yaml/Mon, 01 Jan 0001 00:00:00 +0000https://praxis.fast/ai/dev/examples/configs/time-to-first-token.yaml/Measures the elapsed time from request receipt to the first non-empty SSE body chunk and records a praxis_ai_ttft_seconds Prometheus histogram labeled by modelhttps://praxis.fast/ai/dev/examples/configs/token-ceiling.yaml/Mon, 01 Jan 0001 00:00:00 +0000https://praxis.fast/ai/dev/examples/configs/token-ceiling.yaml/Place token_ceiling after any request translation or prompt enrichment so it evaluates the final provider-bound JSON bodyhttps://praxis.fast/ai/dev/examples/configs/token-counting.yaml/Mon, 01 Jan 0001 00:00:00 +0000https://praxis.fast/ai/dev/examples/configs/token-counting.yaml/Extracts token usage from AI inference responses (streaming and non-streaming) and makes counts available to downstream filters via filter metadata as token.input, token.output, and token.totalhttps://praxis.fast/ai/dev/examples/configs/token-rate-limit.yaml/Mon, 01 Jan 0001 00:00:00 +0000https://praxis.fast/ai/dev/examples/configs/token-rate-limit.yaml/Apply token-aware rate limiting to inference requests and reconcile reported usage.https://praxis.fast/ai/dev/examples/configs/token-rate-limit-header-keys.yaml/Mon, 01 Jan 0001 00:00:00 +0000https://praxis.fast/ai/dev/examples/configs/token-rate-limit-header-keys.yaml/Extends token-rate-limit.yaml with M5 header dimensions (ai#123 / ai#129): each distinct x-tenant-id value gets its own token budget under the same catch-all rulehttps://praxis.fast/ai/dev/examples/configs/token-rate-limit-hierarchy.yaml/Mon, 01 Jan 0001 00:00:00 +0000https://praxis.fast/ai/dev/examples/configs/token-rate-limit-hierarchy.yaml/Optional org, then team, then user budgets on one token_rate_limit admission (ai#125)https://praxis.fast/ai/dev/examples/configs/token-rate-limit-ip-keys.yaml/Mon, 01 Jan 0001 00:00:00 +0000https://praxis.fast/ai/dev/examples/configs/token-rate-limit-ip-keys.yaml/Honor X-Forwarded-For for the IP budget only when a trusted proxy sanitizes it and direct access is blocked; otherwise omit header to key on the TCP peerhttps://praxis.fast/ai/dev/examples/configs/token-rate-limit-mixed-algorithms.yaml/Mon, 01 Jan 0001 00:00:00 +0000https://praxis.fast/ai/dev/examples/configs/token-rate-limit-mixed-algorithms.yaml/Extends token-rate-limit.yaml with per-rule algorithm choice (ai#789 / praxis#551): each rule in rules: independently picks sliding_window or token_bucket, matched by a static header value. team-alpha gets an exact trailing-window budget; team-beta gets a continuously-refilling buckethttps://praxis.fast/ai/dev/examples/configs/token-rate-limit-model-keys.yaml/Mon, 01 Jan 0001 00:00:00 +0000https://praxis.fast/ai/dev/examples/configs/token-rate-limit-model-keys.yaml/Extends token-rate-limit.yaml with M5 model dimensions (ai#123): each distinct model identity gets its own token budget under the same catch-all rulehttps://praxis.fast/ai/dev/examples/configs/token-rate-limit-soft-enforcement.yaml/Mon, 01 Jan 0001 00:00:00 +0000https://praxis.fast/ai/dev/examples/configs/token-rate-limit-soft-enforcement.yaml/Per-rule enforcement chooses what happens when the algorithm denies a reservation because the token budget is exhaustedhttps://praxis.fast/ai/dev/examples/configs/token-rate-limit-soft-tiers.yaml/Mon, 01 Jan 0001 00:00:00 +0000https://praxis.fast/ai/dev/examples/configs/token-rate-limit-soft-tiers.yaml/Extends token-rate-limit.yaml with graduated enforcement tiers (proposal S1, ai#881)https://praxis.fast/ai/dev/examples/configs/token-usage-headers.yaml/Mon, 01 Jan 0001 00:00:00 +0000https://praxis.fast/ai/dev/examples/configs/token-usage-headers.yaml/Inject Praxis-Token-Input, Praxis-Token-Output, and Praxis-Token-Total headers into downstream responses when token counts are available in filter metadata