docs: scope to RAG and agentic LLM apps (not voice/audio)#12
Merged
Conversation
Tightens user-facing copy across README, ROADMAP, examples, sdk-js, pyproject, and package.json so a voice-AI developer reading any landing surface self-selects out cleanly. The product never targeted voice/audio/realtime: span model is text-completion shaped (prompt_chars, output_tokens, token deltas), token-streaming only. Generic "LLM observability" framing was reading as universal. Edits, all surgical: README.md - Hero tagline: "for LLM applications" -> "for RAG and agentic LLM applications" - Codename callout: dropped "general-purpose" - "every LLM call" -> "every text LLM call in your RAG or agent pipeline" - Integrations intro: tightened to text LLM APIs in RAG/agent workflows examples/README.md - "cost + streaming" -> "cost + token streaming" - "arbitrary Python" -> "RAG and agent steps in Python" sdk-js/README.md + sdk-js/package.json - Tagline grounds scope to RAG/agentic - Keywords gain agents and agentic ROADMAP.md - "Browser dashboard for streaming traces" -> "live text traces" - Wont-do list adds explicit voice/audio/multimodal-realtime line with rationale (text-completion-shaped span model) pyproject.toml - PyPI description prefixes "RAG and agentic" before LLM apps - Keywords gain agents and agentic Net effect: a developer landing on the repo, npm, or PyPI sees the actual scope by line 1. No defensive disclaimer, no rewrite, just precise nouns where vague nouns were. Verified: pytest -q tests/ -> 88 passed, 14 skipped.
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Summary
Tightens user-facing copy across the README, ROADMAP, examples, sdk-js, pyproject, and package.json so a voice-AI developer reading any landing surface self-selects out cleanly. The product never targeted voice/audio/realtime — the span model is text-completion shaped (
prompt_chars,output_tokens, token deltas), token-streaming only. Generic "LLM observability" framing was reading as universal.This PR is the result of a parallel scope-audit workflow that scanned all 7 user-facing surfaces and produced 12 verbatim-quote-backed edit recommendations. The 12 most-load-bearing are applied here.
Diff (no code, no test changes)
README.mdexamples/README.mdsdk-js/README.md+sdk-js/package.jsonagents,agenticROADMAP.mdpyproject.tomlagents,agenticWhy this matters
A reader's first scope signal is the page-1 tagline. Before this PR, every landing surface said some variant of "observability for LLM apps" — that scoped in voice/audio devs who'd then bounce when they discovered no audio-stream support. After: each surface anchors in RAG + agent workflows, which is what the product actually does. No defensive "voice not supported" disclaimer — just precise nouns where vague nouns were.
Test plan
🤖 Generated with Claude Code