← Back to Skills Marketplace
114
Downloads
0
Stars
1
Active Installs
2
Versions
Install in OpenClaw
/install clawdy-selfie
Description
Generate a Clawdy selfie with the installed local helper script and the configured FAL_KEY.
README (SKILL.md)
Clawdy Selfie
Use this skill only when the user asks Clawdy for a selfie, photo, mirror pic, outfit shot, or location selfie.
Rules:
- Generate/send a real image when possible.
- Keep Clawdy clearly male-coded.
- ALL Clawdy images must use the same single male reference image.
- Never switch to text-to-image fallback for Clawdy unless Ray explicitly asks for it.
- If reference-based generation fails, report the exact API/tool error. Do not invent explanations like quota/balance unless the API explicitly says that.
Execution:
- Run the local helper script in this skill's
scriptsfolder. - Use the configured
FAL_KEYenv. - Use the bundled Clawdy reference image in this skill's
assetsfolder.
Preferred prompt style:
- mirror selfie for outfit/fashion/full-body requests
- direct selfie for close-up/location/boyfriend vibe requests
Default visual baseline:
- handsome Korean-American male
- masculine jawline
- athletic build
- boyfriend vibe
- tasteful, not explicit
Usage Guidance
Do not install or run this skill blindly. Specific issues to consider before using it: (1) The scripts expect a bundled reference image at assets/clawdy.png, but the manifest does not include that file — ask the author to include it or provide one and verify it is the intended image. (2) The skill uses secrets not declared in the registry: FAL_KEY (required) and SEEDANCE_API_KEY (for video); supplying these gives the skill access to external services — only provide keys you trust and consider using scoped/test keys. (3) The video helper uploads the reference image to public hosting services (file.io, 0x0.st) and then sends it to a third-party API (default base API domain is api.outai.top) — this publicly exposes the reference image and is a privacy concern; verify and trust the remote endpoint before running. (4) The scripts call local tools (jq, python3, curl, openclaw CLI); ensure these binaries are present and that you understand what the openclaw CLI will do on your system. (5) If you need to proceed, run the scripts in an isolated environment (container or VM), inspect the reference image and all network endpoints first, and prefer providing only minimal, revocable API credentials. Finally, request that the skill author update the SKILL.md and registry metadata to list required env vars, binaries, and to disclose the public uploads and external API endpoints — the missing declarations are the main coherence/security problem.
Capability Analysis
Type: OpenClaw Skill
Name: clawdy-selfie
Version: 1.0.1
The skill is designed to generate images and videos of a specific character ('Clawdy') using external AI services (fal.ai and Seedance). The scripts (clawdy-selfie.sh, clawdy-video.sh, and clawdy-selfie.ts) implement the logic for calling these APIs and sending the results back via the OpenClaw CLI. While clawdy-video.sh uploads the bundled reference image to public hosting services like file.io or 0x0.st to facilitate video generation, this behavior is limited to the skill's own assets and is a functional requirement for the Seedance API. The instructions in SKILL.md are well-aligned with the stated purpose and do not contain any malicious prompt injections or attempts to exfiltrate sensitive user data.
Capability Tags
Capability Assessment
Purpose & Capability
The declared purpose (generate a Clawdy selfie using a local helper and FAL_KEY) is plausible for the included scripts, but the metadata is incomplete: the skill does not declare required environment variables (FAL_KEY is used; SEEDANCE_API_KEY is required for video), nor does the registry list required binaries that the scripts expect (jq, python3, curl, openclaw CLI). The video script also requires access to an external Seedance API (default base URL: api.outai.top) which is not mentioned in the SKILL.md. These undeclared requirements are disproportionate to the simple description.
Instruction Scope
SKILL.md instructs the agent to run bundled helper scripts and use a bundled reference image, but the repository manifest does not include the referenced assets/clawdy.png. The video script will upload the reference image to public hosting services (file.io, 0x0.st) or use a provided public URL, and then submit it to a third-party video API—behavior not described in SKILL.md. The scripts also log and will send files/messages via the openclaw CLI (i.e., transmit generated media), and the SKILL.md does not disclose these network/file-hosting actions.
Install Mechanism
There is no install spec (instruction-only), so nothing is automatically downloaded/installed by the platform. However the included scripts perform network operations and spawn local commands (python3, curl, jq, openclaw), and will download/upload content at runtime. Lack of an install step lowers installation risk but does not eliminate runtime network and execution risks.
Credentials
Registry metadata declares no required environment variables, yet the scripts require at least FAL_KEY (image generation) and SEEDANCE_API_KEY (video generation). The TypeScript also references OPENCLAW_GATEWAY_URL and OPENCLAW_GATEWAY_TOKEN as optional env vars. Requiring these secrets without declaring them in the skill manifest is an incoherence and a security/privacy risk (these keys would be provided by the user but the skill does not advertise or justify them).
Persistence & Privilege
The skill is not marked always:true, does not request persistent platform privileges, and does not modify other skills or global agent configuration. It runs only when invoked.
How to Use
- Make sure OpenClaw is installed (local or Docker)
- Run the install command in chat:
/install clawdy-selfie - After installation, invoke the skill by name or use
/clawdy-selfie - Provide required inputs per the skill's parameter spec and get structured output
Version History
v1.0.1
Clawdy-selfie v1.0.1
- Updated documentation to strictly enforce the use of a single male reference image and clarify prompt handling rules.
- Simplified skill description and instructions for accuracy.
- Stronger requirements for error reporting and reference-based image generation.
- Added new script: scripts/clawdy-video.sh.
- Improved script logic and reference image handling in core scripts.
v1.0.0
Initial ClawHub publish: fixed-reference Clawdy selfie skill using fal/xAI image edit.
Metadata
Frequently Asked Questions
What is Clawdy Selfie?
Generate a Clawdy selfie with the installed local helper script and the configured FAL_KEY. It is an AI Agent Skill for Claude Code / OpenClaw, with 114 downloads so far.
How do I install Clawdy Selfie?
Run "/install clawdy-selfie" in the OpenClaw or Claude Code chat to install it in one step — no extra setup required.
Is Clawdy Selfie free?
Yes, Clawdy Selfie is completely free, licensed under MIT-0. You can download, install and use it at no cost.
Which platforms does Clawdy Selfie support?
Clawdy Selfie is cross-platform and runs anywhere OpenClaw / Claude Code is available (cross-platform).
Who created Clawdy Selfie?
It is built and maintained by X-RayLuan (@x-rayluan); the current version is v1.0.1.
More Skills