Update AssemblyAI skill: drop gemini-3-flash-preview, spoken-digit tool pattern trap, granular location PII policies#18
Open
dlange-aai wants to merge 1 commit into
Conversation
…git tool pattern trap + granular location PII policies - LLM Gateway: drop retired gemini-3-flash-preview model (removed from spec #198) - Voice Agents: document tool parameter pattern/examples hints for spoken digit sequences and the re-ask-loop trap from over-strict patterns (#208) - API reference: explain shape-dependent matching of granular location_* PII redaction policies; Entity Detection only returns broad location (#201) - SKILL.md: add Key Gotcha for spoken-digit tool pattern trap
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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
Incremental docs-sync update against the latest assemblyai-docs spec, covering changes landed after the 2026-07-06 run (PR #17).
Changes
references/llm-gateway.md): remove the retiredgemini-3-flash-previewmodel row. It was removed from the available-models list in the spec (assemblyai-docs #198, 2026-07-07). The rest of the model list already matches the current spec.references/voice-agents.md+ aSKILL.mdKey Gotcha): document tool parameterpattern/exampleshints for spoken digit sequences. The agent's cleanup of spoken input is best-effort — long digit sequences (card/account numbers) often arrive with interior spaces, so an over-strictpattern(^\d{16}$) rejects them and traps the caller in a re-ask loop. Added the recommended pattern + handler-side normalization. Source: assemblyai-docs #208, 2026-07-09.references/api-reference.md): explain the shape-dependent matching of the granularlocation_*PII redaction policies (the values were already listed). A combined phrase likeToronto, Canadamatches only the broadlocation, not the granular subtypes; Entity Detection returns only the broadlocation. Source: assemblyai-docs #201, 2026-07-08.Reviewed but no change needed
speech_modelsdefault — the OpenAPIdefaultis the single-element["universal-3-5-pro"], but the user-facing model-selection doc states the effective default is["universal-3-5-pro", "universal-2"], which is what the skill already says. Accurate.language_codes,language_detection, code-switching, 18 languages, three streaming models) that the skill already covers accurately. No new API surface.nanolegacy speech_model removal (#196) — the skill never referencednano.