메뉴
Convert text to natural-sounding speech (MP3) with Amazon Polly and return a shareable HTTPS URL. Use when asked to read something aloud, narrate a document, or make an audio version or podcast of text.
Codex 또는 Claude로 설치 이 Prompt를 복사해 Codex, Claude 또는 다른 어시스턴트에 붙여 넣으면 Skill 페이지를 검토하고 설치를 진행할 수 있습니다.
SOC 직업 분류 기준
Use this to read AI Studio's live capability catalog and repository catalog, execute assistants, and work with decisions. Prefer the caller-bound OAuth connection; current scopes and repository/item/segment ACLs are enforced server-side on every request.
Resolve a district email address or a Google Chat `users/{id}` to the real person in the Peninsula SD directory — name, title, department. Use this whenever you are about to refer to someone by a raw email address or a Chat sender id, instead of guessing who they are.
Schedule an agent task to run later using Amazon EventBridge Scheduler — either a recurring cron/rate, OR a ONE-SHOT follow-up at a specific time. Use this for morning briefs, weekly summaries, reminders, cron jobs, AND ALSO for any "I'll get back to you in N minutes" promise you make to the user. Do NOT use OpenClaw's built-in cron subsystem; it is disabled in this deployment.
Google Workspace (Gmail, Calendar, Drive, Docs, Sheets, Slides, Meet, Chat) as the user's agent account. Use for reading or writing email, calendar events, files, or any Workspace data.
Read-only access to PSD's Redshift data warehouse, authenticated with the caller's Cognito identity. Lists tables, inspects schemas, runs RLS-rewritten SELECT queries, and manages cross-session "lessons" learned about the data. The MCP server enforces all access control — the skill simply forwards JSON-RPC calls with the right bearer token.
Read the caller's Plaud voice recordings — list/search recordings and fetch transcripts and AI summaries. Use for Plaud, voice memos, recorded meetings, "my recordings/notes", transcript or summary of a recording.
| name | psd-tts |
| summary | Turn text into a shareable MP3 with Amazon Polly (generative voices) and return a public HTTPS link — narration, briefings, audio summaries. |
| description | Convert text to natural-sounding speech (MP3) with Amazon Polly and return a shareable HTTPS URL. Use when asked to read something aloud, narrate a document, or make an audio version or podcast of text. |
| allowed-tools | Bash(/opt/agentcore-venv/bin/python3:*) |
Convert text to a natural-sounding MP3 with Amazon Polly, upload it to the workspace
S3 bucket under the public public-images/ prefix, and return an unsigned HTTPS URL the
user can play in any browser (Google Chat renders it as a link). Same delivery model as
psd-image-gen and psd-html-artifact.
Identity. Requires --user <caller-email>. Pass the email verbatim from the
[caller: Name <email>] header of the user turn (it scopes the S3 upload path).
Polly is a standard AWS service (not Bedrock) authenticated by the execution role — no API key, no per-character prompt to the model. Long text is split at sentence boundaries and the MP3 chunks are concatenated automatically.
Text comes from --text, --file, or stdin (safe for long input):
/opt/agentcore-venv/bin/python3 /opt/psd-skills/psd-tts/scripts/synthesize.py --user <email> --text "Good morning. Here is today's briefing."
printf '%s' "$LONG_TEXT" | /opt/agentcore-venv/bin/python3 /opt/psd-skills/psd-tts/scripts/synthesize.py --user <email> --voice Matthew
Options:
| Flag | Default | Description |
|---|---|---|
--voice | Ruth | Polly voice id (see references/voices.md) |
--engine | generative | generative | neural | long-form | standard |
--file | — | Read text from a container file instead of --text |
Returns JSON: { "status": "ok", "url": "...", "s3Key": "public-images/.../<uuid>.mp3", "voice": "Ruth", "engine": "generative", "characters": N, "chunks": N, "sharing": "public-by-link" }.
After the skill returns status: ok, your next chat message MUST contain the bare url
on a line by itself (Google Chat renders it as a playable/downloadable link). Do not wrap
it in Markdown link syntax or bold — bare URL only. The URL is public-by-link and does not
expire; do not presign or re-upload it.
generative (default) is Polly's most natural, expressive engine. Default voice
Ruth (warm en-US female); Matthew is the male counterpart.long-form is tuned for narrating articles/training material (en-US: Danielle,
Gregory, Patrick, Ruth) — reach for it on long documents.neural has the widest voice/language coverage if you need a voice the generative
engine doesn't offer.references/voices.md for the full list before overriding the default.bad_args — missing/invalid --user, or no text supplied.too_large — text exceeds 200,000 characters.upstream_error — Polly rejected the request (e.g. a voice not offered by the chosen
engine) or the owner-bound artifact broker failed. Check the voice/engine pairing in
references/voices.md.misconfigured — the trusted web broker is not configured.