docs(anthropic-messages,maas-ipp): configurable vLLM endpoint + customer narrative - #14
Open
jordigilh wants to merge 1 commit into
Open
docs(anthropic-messages,maas-ipp): configurable vLLM endpoint + customer narrative#14jordigilh wants to merge 1 commit into
jordigilh wants to merge 1 commit into
Conversation
anthropic-messages (demos#10): replace the hardcoded vLLM endpoint IP with a configurable VLLM_ENDPOINT env var, consistent with how skillberry-agent-proxy parameterizes environment-specific values -- passthrough.yaml/transform.yaml become .tmpl files expanded via envsubst at run time (git-ignored once rendered), rather than files a demo runner has to hand-edit. maas-ipp (demos#11): add a customer-facing "What this demo proves" section and narrate the existing "Call models" curl steps. Grounded in the actual MAAS_IPP_PROFILE switch (overlays/xks vs overlays/xks-praxis, IPP_IMAGE vs PRAXIS_EXTPROC_IMAGE) and Praxis's documented Pre-Auth/Post-Auth IPP contract: this demo proves Praxis can sit in MaaS's existing Envoy ExternalProcessor slot and preserve IPP's pre/post-auth behavior, not that it introduces a new datapath -- that's the actual customer/roadmap-relevant claim, not just "the curl works." Documentation/config only, no code changes -- matches both issues' stated scope. Signed-off-by: Jordi Gil <jgil@redhat.com>
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
What
Closes #10 and #11.
anthropic-messages(#10):passthrough.yaml/transform.yamlhardcoded a vLLM endpoint IP, forcing a manual edit of two files to run the demo against a different backend. They're now.tmplfiles with"${VLLM_ENDPOINT}", expanded viaenvsubst(git-ignored once rendered) — the same patternskillberry-agent-proxyalready uses for environment-specific values in this repo.maas-ipp(#11): the README was thorough on Forge/ops mechanics but had no "here's what you're about to see and why it matters" framing. Added a What this demo proves section and narrated the existingCall modelscurl steps, grounded in the actualMAAS_IPP_PROFILEswitch (overlays/xks+IPP_IMAGEvs.overlays/xks-praxis+PRAXIS_EXTPROC_IMAGE) and Praxis's Pre-Auth/Post-Auth IPP contract — the point being that this demo proves Praxis can occupy MaaS's existing EnvoyExternalProcessorslot and preserve IPP's behavior, not that it's a new/separate datapath.Verification
.tmplfiles withenvsubstlocally; diffed against the prior hardcoded YAML — byte-identical whenVLLM_ENDPOINTmatches the old default (10.0.0.99:8000).DCO sign-off included.