AI Image Generation and Editing
zhayujie/CowAgent
Generates or edits images from text prompts through a Python script that picks an image backend based on which API keys are configured.
Qwen AI Platform payment skill, used for account balance enquiries, fix-amount recharge, recharge page guidence, TokenPlan quotations, TokenPlan purchases and TokenPlan page guidence.
$ npx skills add QianWen-AI/qianwen-ai --skill qianwen-payment -a claude-codeProject install by default; add -g for ~/.claude/skills/.
$ gh skill install QianWen-AI/qianwen-ai qianwen-payment --agent claude-codeProject scope by default; add --scope user for a personal install. Needs GitHub CLI 2.90.0 or later (public preview).
$ git clone --depth 1 https://github.com/QianWen-AI/qianwen-ai.git skills-src && mkdir -p .claude/skills && cp -r skills-src/skills/ops/qianwen-payment .claude/skills/qianwen-payment && rm -rf skills-srcUse ~/.claude/skills/ instead of .claude/skills for a personal install. The folder must contain SKILL.md.
Claude Code skills documentation · loads skills from .claude/skills/
Install the "qianwen-payment" agent skill from https://github.com/QianWen-AI/qianwen-ai/tree/main/skills/ops/qianwen-payment into .claude/skills/qianwen-payment/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "qianwen-payment", then confirm the skill loads.Claude Code copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
$skill-installer install https://github.com/QianWen-AI/qianwen-ai/tree/main/skills/ops/qianwen-paymentType this inside Codex. $skill-installer <name> installs a curated skill from openai/skills. The installer writes to $CODEX_HOME/skills (default ~/.codex/skills). Restart Codex if the skill does not show up.
$ npx skills add QianWen-AI/qianwen-ai --skill qianwen-payment -a codexProject install goes to .agents/skills/; add -g for ~/.codex/skills/.
$ gh skill install QianWen-AI/qianwen-ai qianwen-payment --agent codexProject scope by default (.agents/skills/); add --scope user for a personal install.
$ git clone --depth 1 https://github.com/QianWen-AI/qianwen-ai.git skills-src && mkdir -p .agents/skills && cp -r skills-src/skills/ops/qianwen-payment .agents/skills/qianwen-payment && rm -rf skills-srcUse ~/.agents/skills/ instead of .agents/skills for a personal install.
Codex skills documentation · loads skills from .agents/skills/
Install the "qianwen-payment" agent skill from https://github.com/QianWen-AI/qianwen-ai/tree/main/skills/ops/qianwen-payment into .agents/skills/qianwen-payment/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "qianwen-payment", then confirm the skill loads.Codex copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
$ npx skills add QianWen-AI/qianwen-ai --skill qianwen-payment -a cursorProject install goes to .agents/skills/; add -g for ~/.cursor/skills/.
$ gh skill install QianWen-AI/qianwen-ai qianwen-payment --agent cursorProject scope by default (.agents/skills/); add --scope user for a personal install.
$ git clone --depth 1 https://github.com/QianWen-AI/qianwen-ai.git skills-src && mkdir -p .cursor/skills && cp -r skills-src/skills/ops/qianwen-payment .cursor/skills/qianwen-payment && rm -rf skills-srcUse ~/.cursor/skills/ instead of .cursor/skills for a personal install.
Cursor skills documentation · loads skills from .cursor/skills/, .agents/skills/, .claude/skills/, .codex/skills/
Install the "qianwen-payment" agent skill from https://github.com/QianWen-AI/qianwen-ai/tree/main/skills/ops/qianwen-payment into .cursor/skills/qianwen-payment/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "qianwen-payment", then confirm the skill loads.Cursor copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
$ gemini skills install https://github.com/QianWen-AI/qianwen-ai.git --path skills/ops/qianwen-payment--scope user (default) or --scope workspace; --path is the subfolder of the repo that holds the skill; --consent skips the security confirmation prompt.
$ npx skills add QianWen-AI/qianwen-ai --skill qianwen-payment -a gemini-cliProject install goes to .agents/skills/; add -g for ~/.gemini/skills/.
$ gh skill install QianWen-AI/qianwen-ai qianwen-payment --agent gemini-cliProject scope by default (.agents/skills/); add --scope user for a personal install.
$ git clone --depth 1 https://github.com/QianWen-AI/qianwen-ai.git skills-src && mkdir -p .gemini/skills && cp -r skills-src/skills/ops/qianwen-payment .gemini/skills/qianwen-payment && rm -rf skills-srcUse ~/.gemini/skills/ instead of .gemini/skills for a personal install, then run /skills reload.
Gemini CLI skills documentation · loads skills from .gemini/skills/, .agents/skills/
Install the "qianwen-payment" agent skill from https://github.com/QianWen-AI/qianwen-ai/tree/main/skills/ops/qianwen-payment into .gemini/skills/qianwen-payment/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "qianwen-payment", then confirm the skill loads.Gemini CLI copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
$ gh skill install QianWen-AI/qianwen-ai qianwen-paymentInstalls for Copilot at project scope by default; add --scope user for a personal install. Preview a skill first with gh skill preview. Needs GitHub CLI 2.90.0 or later (public preview).
$ npx skills add QianWen-AI/qianwen-ai --skill qianwen-payment -a github-copilotProject install goes to .agents/skills/; add -g for ~/.copilot/skills/.
$ git clone --depth 1 https://github.com/QianWen-AI/qianwen-ai.git skills-src && mkdir -p .github/skills && cp -r skills-src/skills/ops/qianwen-payment .github/skills/qianwen-payment && rm -rf skills-srcUse ~/.copilot/skills/ instead of .github/skills for a personal install. Commit .github/skills so cloud agent and code review can use it.
GitHub Copilot skills documentation · loads skills from .github/skills/, .claude/skills/, .agents/skills/
Install the "qianwen-payment" agent skill from https://github.com/QianWen-AI/qianwen-ai/tree/main/skills/ops/qianwen-payment into .github/skills/qianwen-payment/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "qianwen-payment", then confirm the skill loads.GitHub Copilot copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
$ npx skills add QianWen-AI/qianwen-ai --skill qianwen-payment -a opencodeOpenCode documents no install command of its own. Project install goes to .agents/skills/; add -g for ~/.config/opencode/skills/.
$ gh skill install QianWen-AI/qianwen-ai qianwen-payment --agent opencodeProject scope by default (.agents/skills/); add --scope user for a personal install.
$ git clone --depth 1 https://github.com/QianWen-AI/qianwen-ai.git skills-src && mkdir -p .opencode/skills && cp -r skills-src/skills/ops/qianwen-payment .opencode/skills/qianwen-payment && rm -rf skills-srcUse ~/.config/opencode/skills/ instead of .opencode/skills for a personal install.
OpenCode skills documentation · loads skills from .opencode/skills/, .claude/skills/, .agents/skills/
Install the "qianwen-payment" agent skill from https://github.com/QianWen-AI/qianwen-ai/tree/main/skills/ops/qianwen-payment into .opencode/skills/qianwen-payment/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "qianwen-payment", then confirm the skill loads.OpenCode copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
qianwen-paymentQwen AI Platform payment skill, used for account balance enquiries, fix-amount recharge, recharge page guidence, TokenPlan quotations, TokenPlan purchases and TokenPlan page guidence.
Qianwen Payment is an agent skill from QianWen-AI/qianwen-ai. Qwen AI Platform payment skill, used for account balance enquiries, fix-amount recharge, recharge page guidence, TokenPlan quotations, TokenPlan purchases and TokenPlan page guidence. TRIGGER for explicit Qwen AI Platform balance or recharge requests, recharge result/history/page requests, Token Plan quotes or new purchases, or a model error that explicitly reports insufficient cash balance or an overdue account. DO NOT TRIGGER for usage, quota, cost, settled bills, standalone subscription status, or team seats…
Its SKILL.md is about 4.3k tokens, which your agent loads only when the skill is triggered. The skill folder holds 13 other files, including scripts and reference files (for example `references/cli-contracts.md`, `references/recharge-flow.md` and `references/sources.md`).
It works with Qwen. The repository describes itself as: 将 QianWen 的全套 AI 技能折叠进你的 Agent 里,你提需求就行。 The licence is Apache-2.0.
3 steps, taken from the step headings in SKILL.md.
Read from SKILL.md and the folder at commit 11b1805. It shows what the files ask for, not the result of running them.
Pre-approves nothing: there is no allowed-tools line, so your agent's usual permission prompts apply.
From allowed-tools in the SKILL.md frontmatter.
Ships 7 files in scripts/ (JavaScript), which the agent can run.
Shell commands in SKILL.md call:
nodeFrom the folder's file list and the shell code blocks in SKILL.md.
No URLs in SKILL.md.
From URLs in SKILL.md, links to its own repository left out.
Names no API keys, tokens, secrets or passwords.
From names ending in _API_KEY, _TOKEN, _SECRET, _KEY or _PASSWORD in SKILL.md.
Qianwen Payment loads about 4.3k tokens when it runs, and up to ~19k if it reads all its reference files. Until then it costs about 171 tokens; SKILL.md has 2,107 words of instructions outside code blocks.
Estimates: characters ÷ 4, the usual rule of thumb; real counts depend on the model's tokenizer. Scripts and assets cost tokens only if the agent reads them.
The automated check found no risky patterns in SKILL.md.
Automated static check — not a guarantee. Review scripts before installing. It scans the text of SKILL.md for risky patterns (piping downloads into a shell, reading credential files, hidden Unicode, destructive commands); the scripts in this folder are not scanned.
The full file from QianWen-AI/qianwen-ai at commit 11b1805, republished under its Apache-2.0 licence (© QianWen-AI). 2,107 words, ~4,334 tokens.
.claude/skills/qianwen-payment/SKILL.md (or your agent's skills folder). This skill also uses 11 other files; get the full folder from GitHub.Qwen AI Platform (千问AI平台) cash balance, fixed-amount Alipay recharge, and new Token Plan purchase orchestration. Use this Skill only for a Qwen AI Platform account. Do not infer recharge intent or an overdue account from a low balance, a generic quota error, HTTP status, or unrelated model failure.
Recharge and Token Plan are isolated domains. Recharge uses only the original recharge references and scripts listed below. Token Plan uses only its explicitly listed Token Plan files, plus the unchanged generic command-resolver.mjs. Never use one domain's validator, poller, recovery state, or user-facing result wording in the other domain.
Requires Node 18.18+ for every route in this Skill. Recharge uses scripts/preflight.mjs; Token Plan uses scripts/tokenplan-preflight.mjs. Each domain script exclusively owns its runtime baseline, QianWen CLI compatibility, validation, and Alipay path decisions; do not reproduce or override those decisions in prose.
| Location | Authoritative responsibility |
|---|---|
scripts/preflight.mjs | Route dependency probe, amount normalization, and transaction-critical response validation |
scripts/recharge-result-contract.mjs | Recharge-result identity and status classification |
scripts/poll-recharge.mjs | Foreground post-payment result polling |
references/recharge-flow.md | Confirmed-order business sequence, branching, stop and recovery |
references/cli-contracts.md | Commands, arguments, JSON shapes, script interfaces and exit codes |
references/sources.md | Manual provenance lookup only |
scripts/tokenplan-preflight.mjs | Token Plan dependency probes and response validation |
scripts/tokenplan-order-contracts.mjs | Shared Token Plan payment-result and retained-order history contracts |
scripts/poll-tokenplan-payment.mjs | Token Plan bounded foreground payment wait |
references/tokenplan-flow.md | Complete Token Plan dependency, quote, purchase, settlement, activation, and recovery flow |
Load references only when the current route needs their authoritative detail. For fixed-amount recharge, read recharge-flow.md only after the user confirms order creation. Also read its recovery section when the user explicitly reports that the current order's QR was not displayed and the current order's Alipay materials are still retained. For a Token Plan quote, purchase, or existing-purchase follow-up, load tokenplan-flow.md; do not load Recharge references.
| Request | Route | Business boundary |
|---|---|---|
| Balance | balance-summary | Read-only |
| Confirmed overdue or explicit model-call balance error | Ask whether to recharge | Does not authorize an amount or transaction |
| Fixed-amount Alipay recharge | recharge-order | One create confirmation per attempt, then payment and conditional foreground result polling |
| Current-order QR not displayed | missing-qr-recovery | User-triggered presentation recovery from retained Alipay materials; no command or new order |
| Existing-order result | recharge-result | One read-only snapshot; never starts a polling loop |
| Recharge history | recharge-history | Read-only, default 30 days |
| Recharge page guidance | recharge-page | Confirm before opening the browser |
| Token Plan quote or new purchase | token-plan | Follow tokenplan-flow.md; one confirmation authorizes one create request |
| Existing Token Plan purchase follow-up | token-plan-recovery | Query the retained order first; only an explicit request may reuse its materials or resume the same Alipay handoff while it remains unpaid; never creates an order |
direct or alipay-payment-skill before order creation. alipay-payment-skill owns safety, commands, retries, output, and stop decisions only within the delegated payment-side step. Never switch paths after a direct Alipay command has started because its side effects may be unknown.paymentUrl from the current create response. Never accept, rewrite, reconstruct, normalize, decode/re-encode, or substitute a link from the user, history, logs, or another order.category=credited proves the recharge was credited. An Alipay result, QR scan, user statement, balance change, or history entry does not.poll-recharge.mjs. A command result visible only to the Agent, file read, preview, draft, or planned final reply is not delivery.This section applies only to the direct path. The delegated path follows alipay-payment-skill. sessionId is optional and applies only to alipay-bot submit-payment; trigger-payment-signal never receives it. Use the first non-empty value from this order:
| Priority | Source |
|---|---|
| 1 | Runtime metadata: sessionId / session_id |
| 2 | Current conversation: threadId / thread_id / conversationId / conversation_id / agentContextId; for Codex: CODEX_THREAD_ID |
| 3 | Environment variable AIPAY_SESSION_ID |
Use the value as-is. Do not generate one, ask the user for one, or substitute another business ID. If all sources are empty, omit --session-id.
Use only the amount intentionally supplied for the current recharge.
Validate the normalized amount together with route dependencies:
node <skill-dir>/scripts/preflight.mjs --route recharge-order --amount <normalized>For every recharge-order attempt, the compatible Alipay CLI and alipay-payment-skill are independent blocking dependencies. Preflight validates the CLI contract; separately verify that alipay-payment-skill is present and loadable in the host's available-Skill registry. Neither dependency substitutes for the other, even when preflight selects payment_path=direct. Complete both checks before authentication, confirmation, or order creation.
After both dependency gates pass, retain the returned payment_path. direct means the optional pre-handoff signal capability is available; alipay-payment-skill means payment submission must use the already-verified alipay-payment-skill. If either dependency is unavailable, stop and use cli-contracts.md § Dependency setup only when installation guidance is needed.
Verify a server-authenticated QianWen CLI account using the authentication contract in cli-contracts.md. Use only the CLI OAuth device flow; show only its returned verification URL and never ask for credentials or tokens.
Retain the full account identifier only for current-flow consistency checks and show only its masked form. Immediately before creation, re-read the authenticated account and compare it with the confirmed snapshot. Any account, amount, currency, payment-method, or operation-scope change invalidates the confirmation.
Do not query balance, recharge history, or prior orders before creation. They add latency without protecting this transaction.
For every fresh fixed-amount create confirmation, the first visible line must be this complete AI risk notice, localized when appropriate:
AI risk notice: This recharge request was prepared by AI and may not reflect your actual intent. Review the amount, account, and operation scope; do not confirm if anything is incorrect.Then show the masked account, normalized amount, Alipay method, and that confirmation creates one recharge order and enters the payment flow. End with a natural confirmation question. Do not use a mechanical Confirm/Cancel option list.
Recharge canceled.Communicate only these business transitions. Do not expose CLI commands, arguments, skill routing, validators, service names, polling internals, sessionId, the internal polling deadline, or raw QianWen-side errors unless the user explicitly asks for troubleshooting information.
After confirmation and before invoking the create command:
Creating the recharge order…MEDIA: <path-or-url> lines, extract every reference and remove only those lines from the text. Send the remaining text unchanged, then send the referenced images in their original order through the image/media channel.alipay-payment-skill without reproducing them here. The scope of its lifecycle terminators within this outer flow is defined in recharge-flow.md. If it already emitted content through a user-visible channel, do not repeat it.For the direct path, MEDIA: handling changes only the transport representation and is not Alipay result validation. For either path, Progress node 2 is complete only after the selected path's content is user-visible. If the runtime cannot publish that text without ending the outer turn, use it as the final reply, do not poll in that turn, and use one existing-order snapshot after the user's next message. Never defer the Alipay content into a planned final reply after polling. The pass-through rule overrides the QianWen-side localization, hiding, and error-rewriting rules above.
If the confirmed flow requests no-output recovery because the payment step produced no user-facing content, show the localized equivalent of: The payment step did not return any displayable result, so its outcome is unknown. The existing recharge order has been retained. Do not create another order or pay again; ask me to check this order before retrying. This is a QianWen-owned recovery notice, not Alipay output or proof of the recharge result.
The exactly-once rule has one exception: when the user reports that the current payment QR was not displayed, follow recharge-flow.md § 4 to re-present the QR and payment link from the Alipay data returned for that payment.
Use the category returned by the shared result classifier:
credited: Recharge credited. ¥<amount> has been added to your QianWenAI balance. Then append Recharge records may be delayed by approximately 10 seconds.failed: Recharge was not credited. Check the Alipay transaction status before trying again.processing or unconfirmed: Recharge result not yet confirmed. If you have already paid, do not pay again — tell me once payment is complete and I will check the final result for you.When the confirmed-flow contract requests a post-terminal balance query, append Current available balance: ¥<balance>. after a validated response. If that query fails, append Latest balance is temporarily unavailable. Never alter or replace the recharge result based on the balance response.
Never say that you will report the result automatically or later. The user must send another message before a later snapshot can be queried.
Use cli-contracts.md § Balance. Show the validated CNY amount. A low or zero balance alone does not prove overdue status or authorize recharge.
Use one snapshot from cli-contracts.md § Existing-order result. Do not query Alipay status or start poll-recharge.mjs. Present the result with Progress node 3.
Use cli-contracts.md § Recharge history. On exit 0, convert the CLI
result into a human-readable response. History is informational and never proves a particular order
result.
Explain that the command opens the validated QianWenAI recharge page and ask whether to proceed before using cli-contracts.md § Recharge page guidance. Show the returned validated URL; never construct one.
For a quote, new Individual or Team purchase, payment continuation, missing payment material, or retained-order result check, follow tokenplan-flow.md. This runbook and its scripts exclusively own Token Plan validation, handoff, activation, and recovery.
Do not renew, upgrade, or add quota to a Token Plan. Route standalone existing-plan status, quota, team-seat, and billing questions to qianwen-usage.
For update requests, use the sibling qianwen-update-check Skill when available; otherwise run qianwen version --check. Never run an update check during an unresolved recharge or Token Plan payment flow.
© QianWen-AI, Apache-2.0. Rendered from Markdown: HTML in the file is shown as text, images as links, and headings moved down two levels. Raw file
SKILL.md and 11 other files (scripts, references) in skills/ops/qianwen-payment of QianWen-AI/qianwen-ai.
Open the folder on GitHubat commit 11b1805
Qianwen Payment next to the 5 skills that share the most tags, products or categories with it. Stars are the repository's; “used in” counts other GitHub owners with a copy.
| Skill | Stars | Used in | Tokens | Auto-check | Licence | Repo updated |
|---|---|---|---|---|---|---|
| Qianwen Payment this skillQianWen-AI/qianwen-ai | 105 | — | ~4.3k | Automated safety check: Pass | Apache-2.0 | |
| AI Image Generation and Editingzhayujie/CowAgent | 47k | — | ~1.3k | Automated safety check: Pass | MIT | |
| Agent Feature ReproductionQwenLM/qwen-code | 28k | — | ~1.5k | Automated safety check: Pass | Apache-2.0 | |
| Qwen Code E2E TestingQwenLM/qwen-code | 28k | — | ~2.1k | Automated safety check: Pass | Apache-2.0 | |
| Game Asset Generatorhtdt/godogen | 7.1k | — | ~2.8k | Automated safety check: Pass | MIT | |
| Fix Art IssuesOpenPipe/ART | 11k | — | ~840 | Automated safety check: Notes | Apache-2.0 |
zhayujie/CowAgent
Generates or edits images from text prompts through a Python script that picks an image backend based on which API keys are configured.
QwenLM/qwen-code
Reproduces a feature from Codex or Claude Code in Qwen Code by running the reference agent under capture, reading the traces, then implementing matching behavior.
QwenLM/qwen-code
Guides end-to-end testing of the Qwen Code CLI in headless mode with real model calls, MCP test servers and inspection of raw API traffic.
htdt/godogen
Generates game art from text prompts: PNG images, GLB 3D models, rigged characters, animations and sprites, with background removal.
OpenPipe/ART
Fix a GitHub issue on OpenPipe/ART and open a PR. An agent skill from OpenPipe/ART.
AMAP-ML/LongHorizon-Harness
Reproduce CUA-Harness experiments on WeaveBench from a GitHub checkout.
QianWen-AI/qianwen-ai
Synthesize speech from text with Qwen TTS models. An agent skill from QianWen-AI/qianwen-ai.
QianWen-AI/qianwen-ai
Generate and edit images using Wan and Qwen Image models. An agent skill from QianWen-AI/qianwen-ai.
QianWen-AI/qianwen-ai
Generate text, have conversations, write code, reason, and call functions with Qwen models.
QianWen-AI/qianwen-ai
Generate videos using Wan and HappyHorse models. An agent skill from QianWen-AI/qianwen-ai.
QianWen-AI/qianwen-ai
Understand images and videos with Qwen vision models. An agent skill from QianWen-AI/qianwen-ai.
QianWen-AI/qianwen-ai
Check for qianwen-ai updates and notify the user when a new version is available.
Works with
Qwen AI Platform payment skill, used for account balance enquiries, fix-amount recharge, recharge page guidence, TokenPlan quotations, TokenPlan purchases and TokenPlan page guidence. Qianwen Payment is an agent skill from QianWen-AI/qianwen-ai. Qwen AI Platform payment skill, used for account balance enquiries, fix-amount recharge, recharge page guidence, TokenPlan quotations, TokenPlan purchases and TokenPlan page guidence.
Qianwen Payment fits situations like: explicit Qwen AI Platform balance; recharge requests; recharge result/history/page requests; token Plan quotes.
Run `npx skills add QianWen-AI/qianwen-ai --skill qianwen-payment -a claude-code`. Or copy the skill folder (skills/ops/qianwen-payment in QianWen-AI/qianwen-ai) into .claude/skills/qianwen-payment in your project. Claude Code loads it when a task matches its description.
Run `npx skills add QianWen-AI/qianwen-ai --skill qianwen-payment -a codex`. Or copy the skill folder (skills/ops/qianwen-payment in QianWen-AI/qianwen-ai) into .agents/skills/qianwen-payment in your project. Codex loads it when a task matches its description.
Cursor, Gemini CLI, GitHub Copilot and OpenCode also load SKILL.md folders. With the skills CLI, run `npx skills add QianWen-AI/qianwen-ai --skill qianwen-payment -a cursor` (or -a gemini-cli, github-copilot or opencode for the others). To copy it by hand, put the folder in .cursor/skills/qianwen-payment, .gemini/skills/qianwen-payment, .github/skills/qianwen-payment and .opencode/skills/qianwen-payment in your project.
Going by SKILL.md and its folder, Qianwen Payment needs JavaScript for the scripts in its folder and the command-line tools its instructions call (node). Our summary lists: Node.js.
SKILL.md contains no URLs. Any network use would come from the scripts or tools the agent runs. This is read from the text; nothing was executed.
Our automated static check of SKILL.md found no risky patterns, such as piping downloads into a shell, reading credential files or hidden Unicode. It is not a guarantee. The check reads SKILL.md only: the scripts in the folder are not scanned, so read them before running anything.
Qianwen Payment is published under the Apache-2.0 licence (the repository's licence). It allows redistribution, so the full SKILL.md is shown on this page.
About 4.3k tokens (SKILL.md is roughly 17k characters). Agents keep only the skill's name and description in context until a task matches; then they load SKILL.md in full. Its references folder adds about 15k tokens, read only when the agent opens those files.
Skills that share tags, products or a category with Qianwen Payment: AI Image Generation and Editing (zhayujie/CowAgent, 47k stars), Agent Feature Reproduction (QwenLM/qwen-code, 28k stars), Qwen Code E2E Testing (QwenLM/qwen-code, 28k stars) and Game Asset Generator (htdt/godogen, 7.1k stars). The comparison table on this page puts their stars, adoption, token cost, safety result and licence side by side.
QianWen-AI (a GitHub organization) maintains it in QianWen-AI/qianwen-ai, which has 105 GitHub stars. The repository holds 11 skills in this directory. The repository was last updated on October 10, 2026.
Source: QianWen-AI/qianwen-ai on GitHub. Facts on this page come from the repository at the commit we read; the author's words are quoted as theirs.