https://praxis.fast/ai/dev/Recent content in Praxis AI Documentation on PraxisHugoenhttps://praxis.fast/ai/dev/developing/adding-filters/Mon, 01 Jan 0001 00:00:00 +0000https://praxis.fast/ai/dev/developing/adding-filters/Add a built-in Praxis AI filter by following the filter contract and registering it with the product.https://praxis.fast/ai/dev/architecture/agentic-protocols/Mon, 01 Jan 0001 00:00:00 +0000https://praxis.fast/ai/dev/architecture/agentic-protocols/JSON-RPC 2.0 envelope parsing and protocol-specific metadata extraction for MCP and A2A agent traffic.https://praxis.fast/ai/dev/architecture/ai-inference/Mon, 01 Jan 0001 00:00:00 +0000https://praxis.fast/ai/dev/architecture/ai-inference/Body-aware classification, routing, and enrichment for AI inference traffic, built on the filter pipeline and StreamBuffer body access pattern.https://praxis.fast/ai/dev/anthropic-messages/Mon, 01 Jan 0001 00:00:00 +0000https://praxis.fast/ai/dev/anthropic-messages/Praxis supports the Anthropic Messages API (/v1/messages) through five composable filters. Operators can route, validate JSON envelopes, and transform Anthropic requests to reach any backend.https://praxis.fast/ai/dev/anthropic-messages-replay-test-plan/Mon, 01 Jan 0001 00:00:00 +0000https://praxis.fast/ai/dev/anthropic-messages-replay-test-plan/Defines the Anthropic Messages API request and response shapes Praxis AI should cover across replay, passthrough, and translation tests.https://praxis.fast/ai/dev/capacity-planning/Mon, 01 Jan 0001 00:00:00 +0000https://praxis.fast/ai/dev/capacity-planning/Praxis AI runs on the Praxis proxy core and inherits its file descriptor handling.https://praxis.fast/ai/dev/architecture/cryptographic-inventory/Mon, 01 Jan 0001 00:00:00 +0000https://praxis.fast/ai/dev/architecture/cryptographic-inventory/A human-readable inventory of where the Praxis AI proxy performs cryptography in production, which provider executes each operation, and its FIPS disposition (#1220).https://praxis.fast/ai/dev/features/Mon, 01 Jan 0001 00:00:00 +0000https://praxis.fast/ai/dev/features/Praxis AI extends the Praxis proxy framework with AI-specific filters and provider integrations.https://praxis.fast/ai/dev/fips/Mon, 01 Jan 0001 00:00:00 +0000https://praxis.fast/ai/dev/fips/Praxis AI performs all of its cryptography in the system OpenSSL library.https://praxis.fast/ai/dev/developing/fips/Mon, 01 Jan 0001 00:00:00 +0000https://praxis.fast/ai/dev/developing/fips/Local, reproducible checks that a praxis-ai build is on the path to FIPS 140-3 compliance on Red Hat Enterprise Linux.https://praxis.fast/ai/dev/developing/flow-visualizers/Mon, 01 Jan 0001 00:00:00 +0000https://praxis.fast/ai/dev/developing/flow-visualizers/A flow visualizer is a single, self-contained HTML file that explains one Praxis config as an interactive diagram: the filter pipeline, the clusters and external services it talks to, the…https://praxis.fast/ai/dev/developing/getting-started/Mon, 01 Jan 0001 00:00:00 +0000https://praxis.fast/ai/dev/developing/getting-started/compliance check and Red Hat’s scanner need podman on Linux)https://praxis.fast/ai/dev/developing/llmd-integration-testing/Mon, 01 Jan 0001 00:00:00 +0000https://praxis.fast/ai/dev/developing/llmd-integration-testing/Praxis AI includes an AI-owned ext_proc compatibility layer for llm-d integration.https://praxis.fast/ai/dev/migrating-to-0.2/Mon, 01 Jan 0001 00:00:00 +0000https://praxis.fast/ai/dev/migrating-to-0.2/Version 0.2.0 moves token usage parsing from praxis-ai-apis into the private token usage subsystem in praxis-ai-filters.https://praxis.fast/ai/dev/developing/gpu-nightly-suite/Mon, 01 Jan 0001 00:00:00 +0000https://praxis.fast/ai/dev/developing/gpu-nightly-suite/This page is the single entry point for understanding what our GPU-backed nightly CI runs, what hardware it uses, and where the deeper documentation lives.https://praxis.fast/ai/dev/architecture/responses-retained-payload/Mon, 01 Jan 0001 00:00:00 +0000https://praxis.fast/ai/dev/architecture/responses-retained-payload/The Responses agentic loop can hold several representations of one request at once: the raw JSON body, parsed input, restored history, rewritten model input, tool results, streaming events, and a…https://praxis.fast/ai/dev/architecture/opentelemetry/Mon, 01 Jan 0001 00:00:00 +0000https://praxis.fast/ai/dev/architecture/opentelemetry/Add Praxis AI routing decisions to the OpenTelemetry request traces exported by Praxis core.https://praxis.fast/ai/dev/architecture/outbound-callouts/Mon, 01 Jan 0001 00:00:00 +0000https://praxis.fast/ai/dev/architecture/outbound-callouts/Praxis AI treats an outbound target, its resolved socket addresses, and the credentials attached to the request as one trust decision.https://praxis.fast/ai/dev/architecture/postgres-cryptographic-boundary/Mon, 01 Jan 0001 00:00:00 +0000https://praxis.fast/ai/dev/architecture/postgres-cryptographic-boundary/The openai_responses_store and openai_conversations filters can persist state to PostgreSQL.https://praxis.fast/ai/dev/developing/project-management/Mon, 01 Jan 0001 00:00:00 +0000https://praxis.fast/ai/dev/developing/project-management/All repositories in the praxis-proxy organization use a consistent workflow for planning, prioritizing, and tracking work.https://praxis.fast/ai/dev/quickstart/Mon, 01 Jan 0001 00:00:00 +0000https://praxis.fast/ai/dev/quickstart/Build Praxis AI from source for local development, then route a Responses API request to an AI backend.https://praxis.fast/ai/dev/release/Mon, 01 Jan 0001 00:00:00 +0000https://praxis.fast/ai/dev/release/Praxis AI uses Semantic Versioning. The workspace version is the single source of truth, defined in workspace.package.version in the root Cargo.toml. All workspace crates inherit this version.https://praxis.fast/ai/dev/store/legacy-item-position-repair/Mon, 01 Jan 0001 00:00:00 +0000https://praxis.fast/ai/dev/store/legacy-item-position-repair/Before #570, concurrent item appends could assign the same position to two items in one (tenant_id, conversation_id) scope. The unique index added byhttps://praxis.fast/ai/dev/architecture/response-store/Mon, 01 Jan 0001 00:00:00 +0000https://praxis.fast/ai/dev/architecture/response-store/Durable persistence for OpenAI Responses API responses, enabling retrieval (GET), deletion (DELETE), and input-item pagination across proxy restarts.https://praxis.fast/ai/dev/store/schema-migration/Mon, 01 Jan 0001 00:00:00 +0000https://praxis.fast/ai/dev/store/schema-migration/Replace the disposable v3 response store with a fresh v4 database before starting current releases; startup creates the required schema.https://praxis.fast/ai/dev/architecture/routing-facts/Mon, 01 Jan 0001 00:00:00 +0000https://praxis.fast/ai/dev/architecture/routing-facts/The request facts that routing filters may read, where each one comes from, and the rules that keep them trustworthy.https://praxis.fast/ai/dev/developing/cli-vllm-through-praxis/Mon, 01 Jan 0001 00:00:00 +0000https://praxis.fast/ai/dev/developing/cli-vllm-through-praxis/Connect Codex or Claude Code through a local Praxis AI gateway to inference served by vLLM.https://praxis.fast/ai/dev/developing/type-design/Mon, 01 Jan 0001 00:00:00 +0000https://praxis.fast/ai/dev/developing/type-design/Make invalid states unrepresentable. The type system and serde should enforce constraints at parse time, not at runtime.https://praxis.fast/ai/dev/developing/vllm-qualification/Mon, 01 Jan 0001 00:00:00 +0000https://praxis.fast/ai/dev/developing/vllm-qualification/The existing vLLM Integration workflow runs the complete Responses SDK suite on a real GPU for scheduled runs, run_live_vllm=true dispatches, and the vllm-full-suite PR label.