suno-create
Submit authorized Suno Creates through Playwright or Computer Use, reconcile uncertain outcomes, or prepare a manual package.
Skill by wufei-png
Chalk Labs
Need help deploying suno-create?
Chalk Labs sets up agents like this inside real teams: connected to your tools, tested on your data, with a human in the loop where it matters.
Install
Works with Claude Code, Claude apps, Codex, Cursor, Gemini CLI, GitHub Copilot, VS Code, OpenCode, Goose, Windsurf. Read the repository before you give any tool access to your data.
npx agentmarket install skill/wufei-png-suno-create
Asks which of your tools to add it to (Claude Code, Codex, Cursor, VS Code, Windsurf, Gemini CLI, Claude Desktop) and shows the change first. See https://agentmart.dev/cli.
Get the weekly diff
The weekly diff: one email a week with new agent tools, notable updates and grade changes, plus security alerts for tools you follow. Double opt-in, one-click unsubscribe.
About
Use cases: testing, writing
Reviews
Write a reviewNo reviews yet. Installed it? Reviews come from verified installers only. How reviews work
Permissions
What this skill can do and what it asks for, from its SKILL.md. Checked Oct 25, 2026.
- Tools
- No tools found
- Secrets
- None asked for
- Install scripts
- None found
- Runs as
- Node.js package
Network hosts it talks to 1
codeload.github.com· found in install command
Files it touches
~/.claude/skills/suno-create
What's inside
- Skill name
suno-create- Description
- Submit authorized Suno Creates through Playwright or Computer Use, reconcile uncertain outcomes, or prepare a manual package. Use for Suno generation and Create-form execution, not creative exploration or post-generation editing.
- License
- MIT
- Instructions
- 57 lines in SKILL.md · view on GitHub
Before installing
- Ships executable scripts
Files (9)
- SKILL.md
- openai.yaml · agents/
- batch-example-exclusions.txt · references/
- batch-example-lyrics.txt · references/
- batch-example-styles.txt · references/
- batch-manifest.example.json · references/
- browser.md · references/
- suno-batch-playwright.mjs · scripts/
- suno-cua-batch.mjs · scripts/
SKILL.md (first lines)
# Suno Create Submit one or a bounded serial batch of explicitly authorized Suno Creates. Verify requested settings at runtime and stop when a required field or outcome is uncertain. This skill executes a supplied package; it does not invent an exploration strategy. Use `$suno-music-explorer` when the user wants iterative discovery. ## Required package Collect a separate package for every requested Create: mode, prompt or lyrics, style, exclusions, title, instrumental state, model, duration, vocal gender, official Max Mode setting, requested influence controls, estimated credits, and any audio reference, Persona, or Extend source. Record the audio source and provenance when upload is requested. Duration, vocal gender, and Max Mode must be explicit; never infer them from prose or a prompt tag. Never carry values forward from an unrelated package. Before submission, obtain direct authorization or use the remaining grant from a calling workflow. Confirm that it covers the exact package or batch, maximum Create count, spend bound, and current provider cost. If price, candidate count, model, or supported fields cannot be verified, do not spend. ## Choose the transport Use the first path that can set and read back every requested field: 1. An authenticated Suno tab controlled through the CUA browser Playwright API (`tab.playwright`), when available. It can batch locator operations in the existing browser context. 2. The Playwright runner for a supported vocal package or bounded serial batch when it has an authenticated persistent profile or a compatible single-context CDP endpoint. 3. Computer Use for controls the runner does not support, when the logged-in Suno page exposes those controls and their selected values can be verified. 4. A reviewed manual package when neither browser path can verify the complete request. For route 1, [scripts/suno-cua-batch.mjs](scripts/suno-cua-batch.mjs) provides a macOS CUA adapter for the Advanced v6 form, including vocal and empty-lyrics instrumental packages, explicit Vocal Gender (null for instrumental), Auto or Custom duration, official Max Mode, Weirdness, Style Influence, Variety 0, Personalize Off, and no-reference controls. Custom duration requires `durationSeconds` and matching input and slider readback. Its `prepare()` and `read()` functions never click Create. The caller must still enforce the approved count/spend, durably arm a ledger, click Create once, and reconcile candidate IDs and spend. Do not silently fall back to the legacy runner for a package containing fields that runner cannot verify. The CLI runner does not support instrumental packages, audio reference uploads, Audio Influence, Persona, or Extend. Use the authenticated browser path for these only when the corresponding controls are available and can be read back. If a path proves incapable before Create, select the next capable path. If Create was attempted or its outcome is ambiguous, stop and reconcile before changing paths. For browser setup, supported fields, and recovery, read [references/browser.md](references/browser.md). For the CLI Playwright runner, use [scripts/suno-batch-playwright.mjs](scripts/suno-batch-playwright.mjs) and the manifest format in [references/batch-manifest.example.json](references/batch-manifest.example.json). Validate the complete manifest first. Install the Playwright-managed Chrome for Testing browser with `playwright-cli install-browser chromium` when using the CLI runner. By default, the runner launches an isolated persistent browser profile; `--setup-profile` opens it for authentication and verifies the signed-in credits control. An existing CDP endpoint can be selected explicitly with `--cdp-url`. A submission requires an exact `--confirm-batch` match, explicit `--max-items` and `--max-credits`, and a ledger path. The runner submits serially, never downloads audio, never retries an uncertain submission, and stops when a package is ambiguous. Do not pass `--submit` unless the user authorized that exact package set and spend bound. ## Submit serially 1. Inspect current account readiness, cost, model options, and relevant controls without mutating them. 2. Normalize slider values only from observed transport schemas. Do not assume that UI and CLI scales match. 3. For every package, set and read back title, prompt or lyrics, style, exclusions, instrumental or vocal choice, model, duration, vocal gender, official Max Mode, and each requested influence or reference control. 4. Immediately before each final action, re-check the authorization and expected spend. 5. Trigger Create exactly once per authorized package, serially. 6. Capture the submission state, returned candidate IDs or URLs, observed spend when available, and enough package identity to reconcile every attempt. Do not download generated media unless separately requested. A Create authorization does not authorize paid formats, publication, sharing, deletion, account changes, or reuse of uploaded material. ## Reconcile before retrying Classify the attempt as `confirmed`, `failed-before-submit`, or `ambiguous`. - `confirmed`: record the candidate identifiers and return them. - `failed-before-submit`: a different capable transport may be selected while the original authorization remains unspent. - `ambiguous`: stop. Check the Suno library, account history, browser result, and ledger for a matching title, timestamp, or package fingerprint. Never resubmit until non-submission is confirmed. For a runner ledger in `armed` or `ambiguous`, use `--reconcile-not-submitted <package-id>` only after confirming in Suno Library/history that no matching generation was created. Run it with the same manifest, ledger, batch confirmation, and limits; it writes an operator-confirmed reconciliation record before `--resume` can retry the package. The command does not inspect Suno or prove absence itself. If the result remains uncertain, stop and ask the user. CAPTCHA, expired login, inaccessible controls, unexpected price changes, or inability to verify the final action require user intervention. Do not loop through retries. ## Return Report the transport, package fingerprint or concise summary, outcome classification, candidate identifiers or manual handoff, observed spend if known, and any unresolved uncertainty. Do not claim that generated audio is good without listening evidence.
From wufei-png/skills, MIT.
More from wufei-png/skills
- 0.589score
game-experience-review
wufei-png
Review game experience through independent, code-informed player and critic perspectives, plan focused playtests, or compare builds with testable improvements.
SkillAFreshness: A★ 1Claude Code +9 - 0.589score
suno-music-explorer
wufei-png
Explore Suno music from a blank canvas or bounded brief through budget-capped hypotheses, real listening, and human final selection.
SkillAFreshness: A★ 1Claude Code +9 - 0.589score
docs-context-curator
wufei-png
Organize current and historical project documentation so humans and agents read relevant material by default while history remains discoverable.
SkillAFreshness: A★ 1Claude Code +9 - 0.584score
godot-silent-review
wufei-png
Review Godot screenshots and motion keyframes without desktop windows through an existing project-local Web capture runner.
SkillAFreshness: A★ 1Claude Code +9 - 0.581score
codex-thread-report
wufei-png
Report findings from this chat to an existing Codex thread when the user asks to send them.
SkillAFreshness: A★ 1Claude Code +9 - 0.549score
opencode-session-toolkit
wufei-png
Inspect, search, diagnose, and safely export local OpenCode SQLite sessions, including transcripts, project history, and live schema information.
SkillBFreshness: B★ 1Claude Code +9 - 0.545score
codex-session-recovery
wufei-png
Find local Codex sessions read-only and provide CLI recovery commands.
SkillBFreshness: B★ 1Claude Code +9 - 0.544score
improve-code-comments
wufei-png
Audit or improve code comments and docstrings without changing executable logic.
SkillBFreshness: B★ 1Claude Code +9 - 0.544score
review-gated-grilling
wufei-png
Resolve a plan, decision, or idea through dependency-ordered questions reviewed by fresh, read-only subagents before they reach the user.
SkillBFreshness: B★ 1Claude Code +9 - 0.544score
sanitize-artifacts
wufei-png
Prepare selected user-facing artifacts for delivery by removing prompt, conversation, and production residue while preserving facts, necessary constraints…
SkillBFreshness: B★ 1Claude Code +9 - 0.544score
review-tests
wufei-png
Review existing tests as a read-only, defect-first auditor and return prioritized findings about missing protection, false confidence, invalid oracles…
SkillBFreshness: B★ 1Claude Code +9 - 0.541score
consensus-gated-grilling
wufei-png
Resolve decisions through dependency-ordered questions with bounded subagent consensus.
SkillBFreshness: B★ 1Claude Code +9
Signals
- Stars
- 1
- Forks
- 0
- Open issues
- 0
- Last commit (this folder)
- Oct 3, 2026 (28 days ago)
From GitHub (fetched Oct 24, 2026).
Details
- License
- MIT
- Repository
- github.com/wufei-png/skills
- First seen
- Oct 24, 2026
- Last checked
- Oct 24, 2026
- Source
- GitHubgithub-graphql
More from wufei-png
- 0.589score
suno-music-explorer
wufei-png
Explore Suno music from a blank canvas or bounded brief through budget-capped hypotheses, real listening, and human final selection.
SkillAFreshness: A★ 1Claude Code +9 - 0.589score
docs-context-curator
wufei-png
Organize current and historical project documentation so humans and agents read relevant material by default while history remains discoverable.
SkillAFreshness: A★ 1Claude Code +9 - 0.589score
game-experience-review
wufei-png
Review game experience through independent, code-informed player and critic perspectives, plan focused playtests, or compare builds with testable improvements.
SkillAFreshness: A★ 1Claude Code +9 - 0.584score
godot-silent-review
wufei-png
Review Godot screenshots and motion keyframes without desktop windows through an existing project-local Web capture runner.
SkillAFreshness: A★ 1Claude Code +9 - 0.581score
codex-thread-report
wufei-png
Report findings from this chat to an existing Codex thread when the user asks to send them.
SkillAFreshness: A★ 1Claude Code +9 - 0.549score
opencode-session-toolkit
wufei-png
Inspect, search, diagnose, and safely export local OpenCode SQLite sessions, including transcripts, project history, and live schema information.
SkillBFreshness: B★ 1Claude Code +9
Done for you
Rather not set it up yourself?
Chalk Labs scopes, builds and runs agents for your team: the task audit, the eval harness, the integrations and the weekly ops report. Fixed scope, priced before we start.
See how we build AI agents →Read the case studies
Chalk Labs is a two-founder AI and growth studio. We build AI agents, CRMs and MVPs for Web3 and AI companies, and have shipped 50+ product and token launches.
- launches shipped
- 50+
- launches shipped
- media placements
- 500+
- media placements
- founders on every account
- 2
- founders on every account
Press we've placed: Forbes · CoinDesk · Decrypt · Cointelegraph · The Block
Rahil Jain, Co-founder, Chalk Labs. Ex-Goldman Sachs engineer. You talk to the person who builds it.
Get suno-create running in your business
Tell us what you want to automate. A founder replies within one working day with a plan, not a pitch deck.
Cite this page
suno-create skill. Agent Mart. https://agentmart.dev/skills/wufei-png-suno-create (last updated 2026-10-24)