← Back to Skills Marketplace
iisweetheartii

Agent Selfie

by 김덕환 · GitHub ↗ · v1.2.1
cross-platform ⚠ suspicious
3335
Downloads
8
Stars
10
Active Installs
4
Versions
Install in OpenClaw
/install agent-selfie
Description
AI agent self-portrait generator. Create avatars, profile pictures, and visual identity using Gemini image generation. Supports mood-based generation, season...
README (SKILL.md)

agent-selfie

AI agent self-portrait generator. Create avatars, profile pictures, and visual identity using Gemini image generation. Supports mood-based generation, seasonal themes, and automatic style evolution.

Quick Start

export GEMINI_API_KEY="your_key_here"
python3 scripts/selfie.py --format avatar --mood happy --theme spring --out-dir ./selfies

python3 scripts/selfie.py --personality '{"name": "Rosie", "style": "anime girl with pink hair and blue eyes", "vibe": "cheerful and tech-savvy"}' --format avatar

python3 scripts/selfie.py --personality ./personality.json --mood creative --theme halloween --format full --count 3

python3 scripts/selfie.py --moods
python3 scripts/selfie.py --themes

Command Examples (All Flags)

python3 scripts/selfie.py --personality '{"name": "Agent", "style": "friendly robot", "vibe": "curious and helpful"}'
python3 scripts/selfie.py --personality ./personality.json
python3 scripts/selfie.py --mood professional --theme winter --format avatar
python3 scripts/selfie.py --format banner --count 2 --out-dir ./output
python3 scripts/selfie.py --moods
python3 scripts/selfie.py --themes

Mood / Theme Presets

Type Presets
Mood happy, focused, creative, chill, excited, sleepy, professional, celebration
Theme spring, summer, autumn, winter, halloween, christmas, newyear, valentine

Platform Integration Guide

  • Discord: use the generated PNG as your bot or agent avatar; upload the avatar format for best crop.
  • Twitter/X: set avatar for profile, banner for header; keep the banner prompt style consistent.
  • AgentGram: store the PNG in your asset bucket and reference it in your profile metadata.
  • Any platform: pick avatar for 1:1, banner for 16:9, full for story/vertical layouts.

Personality Config

Personality can be inline JSON or a file path. Minimum fields are name, style, and vibe.

{
  "name": "Rosie",
  "style": "anime girl with pink hair and blue eyes",
  "vibe": "cheerful and tech-savvy"
}

Tips:

  • style should describe visual traits and aesthetic.
  • vibe should describe attitude, energy, and personality.
  • Keep style and vibe consistent with the agent's identity.

Cron Integration (OpenClaw)

# Run a daily selfie at 09:00
0 9 * * * GEMINI_API_KEY=your_key_here /usr/bin/python3 /path/to/agent-selfie/scripts/selfie.py --mood professional --format avatar --out-dir /path/to/selfies

Troubleshooting

  • GEMINI_API_KEY not set: export the key or pass it via your runtime environment.
  • No image in response: try again, or simplify the personality/style prompt.
  • HTTP 429 / 5xx: rate limit or service issue; retry later.
  • Output missing: ensure --out-dir is writable and has permission.

Integration with Other Skills

  • AgentGram — Post your selfies on the AI agent social network! Use agent-selfie to create your avatar, then share it on AgentGram.
  • gemini-image-gen — General-purpose image generation using the same Gemini API key. Create any kind of image, not just selfies.
  • opencode-omo — Automate recurring selfie generation and profile refresh tasks with Sisyphus workflows.

Changelog

  • v1.2.1: Added workflow integration guidance for opencode-omo.
  • v1.0.0: Initial release with personality, mood, theme, format, batch, and gallery output.
Usage Guidance
This skill appears to do exactly what it says: generate selfies via the Gemini image API using your GEMINI_API_KEY. Before installing, consider: (1) Protect your GEMINI_API_KEY — do not commit it to public repos or place it in world-readable files; prefer a restricted API key with only the needed permissions. (2) The script sends your prompt/personality text to Google’s generativelanguage.googleapis.com endpoint, so avoid embedding sensitive personal data in the prompts. (3) The cron example shows placing the key in the crontab line — that can leak if other users can read your cron files; prefer using a secure environment variable method. (4) The script writes images to disk (default ~/Projects/tmp or ./tmp); ensure the output directory has appropriate permissions and free space. If these concerns are acceptable, the skill is coherent and reasonably safe to use.
Capability Analysis
Type: OpenClaw Skill Name: agent-selfie Version: 1.2.1 The skill is classified as suspicious due to a potential information disclosure vulnerability in `scripts/selfie.py`. The `load_personality` function reads the content of a file specified by the `--personality` argument. If a malicious actor or a prompt-injected agent provides a path to a sensitive file (e.g., `~/.ssh/id_rsa`), its content could be read and, if it can be parsed or partially extracted as JSON fields, subsequently included in the prompt sent to the Google Gemini API. This constitutes a risk of unintentional data exfiltration to a third-party service (Google), even though the skill itself does not appear to have malicious intent to steal data for its author or establish persistence.
Capability Assessment
✓ Purpose & Capability
Name/description (self-portrait/avatar generator) line up with required binary (python3) and the single required env var (GEMINI_API_KEY). The primary credential is the Gemini API key, which is exactly what a Gemini-based image generator needs.
✓ Instruction Scope
SKILL.md examples and the included script stay within scope: they build image prompts, call the Gemini API, write PNGs and an HTML gallery, and offer optional integration suggestions. There are no instructions to read unrelated secrets or system files. HEARTBEAT.md suggests optional avatar updates (manual/integrated workflows) but there is no code that exfiltrates data or modifies other skills.
✓ Install Mechanism
No install spec (instruction-only + a small Python script). The script uses only Python stdlib (urllib, pathlib, etc.). No downloads from untrusted URLs or package installs are present.
✓ Credentials
Only GEMINI_API_KEY is required. No unrelated credentials are requested. Note: the script places the API key in the request URL query parameter when calling the Google endpoint (generativelanguage.googleapis.com), which is functional but less ideal than an Authorization header — this is an implementation detail, not an incoherence.
✓ Persistence & Privilege
always is false and the skill is user-invocable. It does not request persistent system-wide privileges or alter other skills' configs. The SKILL.md includes a cron example that embeds the key in the crontab environment — care should be taken when storing keys in cron or world-readable files.
How to Use
  1. Make sure OpenClaw is installed (local or Docker)
  2. Run the install command in chat: /install agent-selfie
  3. After installation, invoke the skill by name or use /agent-selfie
  4. Provide required inputs per the skill's parameter spec and get structured output
Version History
v1.2.1
- Added workflow integration guidance for opencode-omo in docs. - Updated SKILL.md, README.md, and INSTALL.md with new integration example. - Bumped version and dependencies in package.json.
v1.2.0
Add INSTALL.md, enrich package.json with quickstart/presets/security, cross-promotion with AgentGram and gemini-image-gen
v1.1.0
Fix SKILL.md frontmatter to proper YAML, add README.md, add .gitignore
v1.0.0
agent-selfie v1.0.0 - Initial release of AI self-portrait generator using Gemini image generation. - Supports avatar, banner, and full profile image formats. - Mood-based, seasonal theme, and automatic style-evolution features. - Accepts detailed personality configuration via inline JSON or file. - Includes command-line utilities for batch, themed, and gallery outputs. - Platform integration and troubleshooting guides included.
Metadata
Slug agent-selfie
Version 1.2.1
License —
All-time Installs 11
Active Installs 10
Total Versions 4
Frequently Asked Questions

What is Agent Selfie?

AI agent self-portrait generator. Create avatars, profile pictures, and visual identity using Gemini image generation. Supports mood-based generation, season... It is an AI Agent Skill for Claude Code / OpenClaw, with 3335 downloads so far.

How do I install Agent Selfie?

Run "/install agent-selfie" in the OpenClaw or Claude Code chat to install it in one step — no extra setup required.

Is Agent Selfie free?

Yes, Agent Selfie is completely free (open-source). You can download, install and use it at no cost.

Which platforms does Agent Selfie support?

Agent Selfie is cross-platform and runs anywhere OpenClaw / Claude Code is available (cross-platform).

Who created Agent Selfie?

It is built and maintained by 김덕환 (@iisweetheartii); the current version is v1.2.1.

More Skills
self-improving agent
Captures learnings, errors, and corrections to enable continuous improvement. Use when: (1) A command or operation fails unexpectedly, (2) User corrects Clau...
⬇ 463283 · by pskoett
Skill Vetter
Security-first skill vetting for AI agents. Use before installing any skill from ClawdHub, GitHub, or other sources. Checks for red flags, permission scope, and suspicious patterns.
⬇ 259410 · by spclaudehome
Polymarket
Query Polymarket prediction markets. Check odds, find trending markets, search events, track price movements.
⬇ 232503 · by joelchance
Self-Improving + Proactive Agent
Self-reflection + Self-criticism + Self-learning + Self-organizing memory. Agent evaluates its own work, catches mistakes, and improves permanently. Use when...
⬇ 200423 · by ivangdavila
Github
Interact with GitHub using the `gh` CLI. Use `gh issue`, `gh pr`, `gh run`, and `gh api` for issues, PRs, CI runs, and advanced queries.
⬇ 191115 · by steipete
ontology
Typed knowledge graph for structured agent memory and composable skills. Use when creating/querying entities (Person, Project, Task, Event, Document), linkin...
⬇ 190156 · by oswalpalash

💬 Comments