← Back to Skills Marketplace
YumFu
by
TommyYanPS
· GitHub ↗
· v1.7.9
· MIT-0
474
Downloads
1
Stars
1
Active Installs
39
Versions
Install in OpenClaw
/install yumfu
Description
Multiplayer text RPG with 10 playable worlds — play together in Telegram groups! Worlds: 笑傲江湖, Harry Potter, Warrior Cats, F15 Down, 龙虾三国, 倚天屠龙记, Game of Thr...
Usage Guidance
This skill appears to be a bona fide multiplayer RPG, but there are important mismatches and defaults you should review before installing:
- Metadata vs runtime: The registry metadata declares no required environment variables, but the skill expects GEMINI_API_KEY (optional for images), and supports YUMFU_NO_IMAGES and YUMFU_NO_LOGGING. Confirm whether your host provides those env vars and whether you want image/TTS generation.
- Privacy: By default the skill logs session turns and writes session logs and images to ~/clawd/memory/yumfu/ and ~/.openclaw/media/outbound/yumfu/. If you prefer no logs or no network image/TTS calls, set YUMFU_NO_LOGGING=1 and YUMFU_NO_IMAGES=1 (the project documents these flags). Test in a private environment first.
- Operational impact: SKILL.md mandates per-turn image generation and default TTS in group chats. Generating and sending images/audio every turn can be bandwidth-heavy and spammy in group chats — consider disabling images or TTS for groups or confirming GEMINI_API_KEY usage limits.
- Installation: The package contains many scripts but no formal install spec. Ensure the host provides the 'uv' binary and Python runtime the scripts expect before enabling the skill.
- Review sources: If you want stronger assurance, inspect the included scripts (load_game.py, save_game.py, session_logger.py, generate_image.py, generate_turn_tts.py) yourself or run them in an isolated sandbox to confirm they behave as described and respect YUMFU_NO_LOGGING.
If you want minimal exposure, install only for single-user testing, unset GEMINI_API_KEY (or set YUMFU_NO_IMAGES=1), and set YUMFU_NO_LOGGING=1 to avoid external calls and persistent conversation logs while you evaluate.
Capability Analysis
Type: OpenClaw Skill
Name: yumfu
Version: 1.7.9
The skill bundle implements a complex multiplayer RPG engine but exhibits high-risk patterns regarding command execution. Multiple scripts (e.g., build_reentry_context.py, deliver_yumfu_turn.py) and the instructions in SKILL.md rely on constructing shell commands using f-strings and executing them via subprocess or the agent's exec tool. This creates a significant surface for shell injection if variables like user_id or universe are not perfectly sanitized. While the bundle includes a SECURITY.md and SECURITY_ANALYSIS.md acknowledging these risks, the architectural choice to manage state through a large number of external Python scripts called via shell remains inherently risky for an AI agent environment.
Capability Tags
Capability Assessment
Purpose & Capability
The skill's name/description (multiplayer text RPG) align with the included scripts (load/save, session logger, image/TTS generation). Requesting the 'uv' binary is reasonable for running its Python/uv-based scripts. However, the metadata declares no required environment variables while the runtime docs and other files repeatedly reference GEMINI_API_KEY, YUMFU_NO_IMAGES, YUMFU_NO_LOGGING, and YUMFU_USER_ID — an omission in registry metadata that creates an incoherence between declared requirements and actual runtime needs.
Instruction Scope
SKILL.md instructs the agent to load and save local save files (~/.clawd/memory/yumfu/... and similar), to always call session_logger.log_turn (mandatory logging in many sections), to generate images for every game turn and deliver them into group chats, and to produce per-turn TTS by default. These are within the domain of a multiplayer RPG, but the instructions also mandate actions that involve writing logs, calling external image/TTS APIs, and automatically posting media into group chats. The skill references environment variables and local paths that are not declared in the registry metadata, and some sections state 'MUST' semantics (e.g., image generation for every key scene) that may be impractical or privacy-sensitive in some deployments.
Install Mechanism
There is no install spec (instruction-only in the registry), which is low risk on its own — but the skill package contains many scripts and assets (dozens of files) and expects to be run from a local skill path (~/clawd/skills/yumfu). The absence of an explicit install step means the agent/host must already provide the runtime (uv, Python) and places the burden on install-time or host provisioning. This mismatch (lots of code but no install metadata) is notable but not necessarily malicious.
Credentials
The registry lists no required env vars, yet SKILL.md and project docs repeatedly reference GEMINI_API_KEY (for AI images), YUMFU_NO_IMAGES, YUMFU_NO_LOGGING, and YUMFU_USER_ID. Image/TTS generation and session logging involve network calls and writing potentially sensitive conversation logs; those environment variables and privacy controls matter but are not declared. The number of variables referenced is reasonable for the feature set, but the omission from metadata and the 'mandatory' wording for some features (images, TTS, logging) make the requested access ambiguous and potentially disproportionate if a host lacks or refuses those credentials.
Persistence & Privilege
The skill writes persistent saves, backups, session logs, and generated images to the user's home directories (~/clawd/memory/yumfu/, ~/.openclaw/media/outbound/yumfu/), which is expected for a game that persists progress and storybooks. The skill is not marked 'always: true' and does not request elevated platform-wide privileges. However, SKILL.md mandates session logging and automatic delivery of media into group chats by default unless explicitly disabled — behavior that increases persistence and automatic external communication unless the user toggles privacy flags.
How to Use
- Make sure OpenClaw is installed (local or Docker)
- Run the install command in chat:
/install yumfu - After installation, invoke the skill by name or use
/yumfu - Provide required inputs per the skill's parameter spec and get structured output
Version History
v1.7.9
Fix daily evolution delivery-state collision so cron updates send image + TTS again
v1.7.8
Storybook quality upgrade, ending auto-generation, GitHub Pages demo
v1.7.7
Clearer gameplay choices, quieter silent execution, and TTS limited to player-facing story text.
v1.7.6
HTML-first storybook export, scene-bound image+text layout, PDF only after visual confirmation
v1.7.5
Remove roadmap from homepage, promote auto art/audio story/storybook features, and add new F15 Down + Game of Thrones showcase images
v1.7.4
Add YumFu Fun Discord invite link and QR code to README and README.zh homepage
v1.7.3
Fix storybook HTML sharing with embedded images, improve historical image resolution, and document verified PDF export workflow
v1.7.2
Delivery flow, TTS, and Journey to the West style fixes
v1.7.1
Add Stripe support links to README and SKILL.md
v1.7.0
Sengoku Chaos playable world: stronger opening flow, ukiyo-e visual style, maritime merchant rise path, and opening-image-required /yumfu start across worlds.
v1.6.0
Dynamic Daily Evolution: sidecar-safe offline world progress, per-player cron, language-aware re-entry, and stronger single-image-per-turn rules.
v1.5.6
YumFu 1.5.6
- Added a new playable world: Captain Underpants (内裤超人), with corresponding world metadata and triggers.
- Updated SKILL.md to include Captain Underpants world details and new triggers (e.g., 乔治, 哈罗德, Professor Poopypants).
- Refreshed world lists in documentation to reflect the added Captain Underpants adventure.
- Minor metadata and documentation improvements for clarity and consistency.
v1.5.5
YumFu 1.5.5 Changelog
- Added Journey to the West (西游记) as a new playable world; updated triggers and descriptions to reflect 9 total worlds.
- Expanded world/character trigger keywords to include 西游记, 孙悟空, 唐僧, 猪八戒, 牛魔王, 取经, Journey to the West.
- Updated skill description and metadata for improved searchability and world coverage.
v1.5.4
YumFu 1.5.4
- Added new Journey to the West (西游记) world (including world data file)
- Updated skill documentation and roadmap to include Journey to the West as an upcoming playable world
- Improved README and SKILL metadata for world lineup clarity
- Package definition updated for new content and fixes
v1.5.3
YumFu 1.5.3
- Expanded to 8 playable worlds, now including 倚天屠龙记, Game of Thrones, and Lord of the Rings in triggers/description.
- Updated skill description, triggers, and features for accuracy and broader language/world support.
- Documentation refinements in SKILL.md and README.md for clarity and completeness.
v1.5.2
Yumfu 1.5.2
- Added "Lord of the Rings" as a new playable world (previously marked as "Soon").
- Updated world list and status in game and documentation.
- Expanded world triggers to include Lord of the Rings keywords (e.g., "LOTR", "Frodo", "Aragorn", "Gandalf", "指环王", "魔戒").
- Updated instructions and examples in SKILL.md for new world availability.
v1.5.1
YumFu 1.5.1
- Added Game of Thrones world (background, metadata, triggers) in preparation for release
- Updated triggers to support Game of Thrones keywords in all major languages
- Expanded "Available Now" list and world selection details in documentation
- No gameplay changes; this is a content and metadata update to prepare for the new world rollout
v1.5.0
YumFu 1.5.0
- Added support for two new worlds: 龙虾三国 (Three Kingdoms, “Lobster Three Kingdoms”) and 倚天屠龙记 (The Heaven Sword & Dragon Saber), including new storylines and playable roles.
- Expanded world triggers for richer name-based activation (倚天屠龙记, 张无忌, 赵敏, 明教, 龙虾三国, 三国, etc.).
- Updated documentation (SKILL.md) to reflect new worlds, features, and upcoming roadmap.
- Updated world listings for clearer status and availability.
v1.4.9
YumFu 1.4.9 Changelog
- Updated skill description to highlight multiplayer support in Telegram groups.
- Clarified that each player gets their own character in a shared world.
- Emphasized features like PvP duels, team quests, and leaderboards.
- Improved triggers and usage guidance for "play together", group RPG, and natural conversation.
- Expanded world list and group adventure details in documentation.
v1.4.8
YumFu 1.4.8
- Updated documentation in README.md for improved clarity and feature listing.
- No functional code changes; package metadata version bump only.
Metadata
Frequently Asked Questions
What is YumFu?
Multiplayer text RPG with 10 playable worlds — play together in Telegram groups! Worlds: 笑傲江湖, Harry Potter, Warrior Cats, F15 Down, 龙虾三国, 倚天屠龙记, Game of Thr... It is an AI Agent Skill for Claude Code / OpenClaw, with 474 downloads so far.
How do I install YumFu?
Run "/install yumfu" in the OpenClaw or Claude Code chat to install it in one step — no extra setup required.
Is YumFu free?
Yes, YumFu is completely free, licensed under MIT-0. You can download, install and use it at no cost.
Which platforms does YumFu support?
YumFu is cross-platform and runs anywhere OpenClaw / Claude Code is available (cross-platform).
Who created YumFu?
It is built and maintained by TommyYanPS (@yumyumtum); the current version is v1.7.9.
More Skills