Index of every page on this site: llms.txt
All skills
microsoft avatar

/azure-ai

@0d60658
by microsoftmicrosoft/skills3.2k stars
373
Run it yourself

Build and troubleshoot Azure AI Search, Speech, Document Intelligence, Content Understanding, Translator, Content Safety, and OpenAI apps. WHEN: AI Search, vector search, hybrid search, semantic search, indexer, speech-to-text, text-to-speech, fast or batch transcription, Speech SDK logs, session ID, result ID, OCR, document translation, glossary.

Use this Skill: https://skilld.dev/gh/microsoft/skills/azure-ai

Nothing lands on disk. Nothing to clean up.

Fork this Skill

Edit a local copy. It keeps the author and licence.

referencesspeechaudio-without-node.md

≈846 tokens on demand. Your agent reads this file only when SKILL.md points to it.

Speech audio and batch checks without Node.js

Use when Node.js 22+ is unavailable and the user declines installing it. These commands print only labels and header numbers, never audio, transcripts, paths or SAS URLs. The PowerShell lines work in Windows PowerShell 5.1 and PowerShell 7, including ConstrainedLanguage mode.

Checks

  1. Size, container and WAV header (replace <local-audio>):

    f="<local-audio>"; if [ ! -r "$f" ]; then echo read_error; else wc -c < "$f"; sig=$(od -An -tx1 -N12 "$f" | tr -d ' \n'); case "$sig" in 52494646????????57415645*) echo wav; od -An -tu2 -j20 -N4 "$f"; od -An -tu4 -j24 -N4 "$f"; od -An -tu2 -j34 -N2 "$f"; od -An -tx1 -j36 -N4 "$f"; od -An -tu4 -j40 -N4 "$f";; 494433*|ff[ef]*) echo mp3;; 664c6143*) echo flac;; 4f676753*) echo ogg;; 1a45dfa3*) echo webm;; ????????66747970*) echo mp4;; *) echo unknown;; esac; fi
    $p = "<local-audio>"; if (-not (Test-Path -LiteralPath $p -PathType Leaf)) { 'read_error' } else { (Get-Item -LiteralPath $p).Length; $b = @(if ($PSVersionTable.PSVersion.Major -ge 6) { Get-Content -LiteralPath $p -AsByteStream -TotalCount 44 } else { Get-Content -LiteralPath $p -Encoding Byte -TotalCount 44 }); $h = ($b | Select-Object -First 12 | ForEach-Object { '{0:X2}' -f $_ }) -join ''; switch -Regex ($h) { '^52494646.{8}57415645' { 'wav'; "formatTag=$($b[20] + 256*$b[21]) channels=$($b[22] + 256*$b[23]) sampleRate=$($b[24] + 256*$b[25] + 65536*$b[26] + 16777216*$b[27]) bits=$($b[34] + 256*$b[35]) dataAt36=$($b[36] -eq 100 -and $b[37] -eq 97 -and $b[38] -eq 116 -and $b[39] -eq 97) chunkAt36Bytes=$($b[40] + 256*$b[41] + 65536*$b[42] + 16777216*$b[43])"; break } '^(494433|FF[EF])' { 'mp3'; break } '^664C6143' { 'flac'; break } '^4F676753' { 'ogg'; break } '^1A45DFA3' { 'webm'; break } '^.{8}66747970' { 'mp4'; break } default { 'unknown' } } }
    • bash prints, after wav: format tag and channels, sample rate, bits, the 4 bytes at offset 36 (64 61 74 61 = data) and that chunk's size.
    • Format tag 1 = PCM, 3 = IEEE float, 6 = A-law, 7 = mu-law, 65534 = extensible (sub-format unknown here). A non-wav label on a .wav file means the extension does not match the content.
    • If offset 36 is not data (dataAt36=False, often a LIST chunk), the data size and duration are unknown. Otherwise duration ≈ chunk bytes ÷ (sample rate × channels × bits ÷ 8).
  2. Limits: compare size with < 500 MB (Fast Transcription) or <= 1 GB (Batch Transcription); compare duration with < 5 hours (Fast).

  3. Batch failures: List Files has no per-file status. Read only the counts from the downloaded TranscriptionReport file:

    grep -o -E '"(successful|failed)TranscriptionsCount"[[:space:]]*:[[:space:]]*[0-9]+' "<report.json>" || echo read_error_or_not_a_report
    $p = "<report.json>"; if (-not (Test-Path -LiteralPath $p -PathType Leaf)) { 'read_error_or_not_a_report' } else { $m = @(Select-String -LiteralPath $p -Pattern '"(successful|failed)TranscriptionsCount"\s*:\s*\d+' -AllMatches | ForEach-Object { $_.Matches } | ForEach-Object { $_.Value }); if ($m.Count -gt 0) { $m } else { 'read_error_or_not_a_report' } }

Leading silence, non-data chunk layouts and compressed-audio duration are unknown without the Node inspector.

Source: SKILL.md on GitHub

skilld matched fixed text patterns in SKILL.md and file names. Patterns miss obfuscated code.

skilld run checks every file with the same patterns. It asks for approval before it loads a Skill with a behavior marked Needs approval.

1 warning11d4 checks · Risk SAFE
  • Gen Agent Trust Hub11d

    This skill provides tools and guidance for developing and troubleshooting Azure AI services. It includes a local diagnostic script for analyzing Speech SDK logs with robust safety controls, along with comprehensive references for official Azure SDKs and secure authentication practices.

  • Socket11d

    No alerts

  • Snyk11d

    Risk: LOW · No issues

  • Runlayer9mo

    15/15 files flagged

Signed by skilld at 0d60658. This ties the file your Agent reads to that commit on GitHub. It does not review the instructions.

Last checked against GitHub yesterday.

Activeupdated 2 days ago
metadata
{
  "author": "Microsoft",
  "version": "1.2.6"
}

README badge

README badge for microsoft/skills/azure-ai