Openai Image Gen 1.0.1

Batch-generate images via OpenAI Images API. Random prompt sampler + `index.html` gallery.

Security Scan

The skill mostly does what it claims (batch image generation via OpenAI Images API) but has inconsistencies in its instructions/metadata that could lead to executing the wrong script or surprising behavior — exercise caution and verify before running.

Purpose & Capability

The name/description align with the included Python script: it generates prompts and calls the OpenAI Images endpoint. However the registry metadata declares no required env vars/credentials while the SKILL.md and script require an OPENAI_API_KEY. That mismatch is unexpected and should be corrected.

Instruction Scope

SKILL.md tells users to run a hard-coded path (python3 ~/Projects/agent-scripts/skills/openai-image-gen/scripts/gen.py) and to open files under ~/Projects/tmp/..., which is inconsistent with the bundled script location. Running the documented absolute path could execute a different file on a user's machine. Otherwise, the script itself only reads an API key, optional base URL/env vars, writes images and an index.html, and makes network calls only to the API base — there is no obvious data exfiltration, but the path/usage inconsistencies are risky.

Install Mechanism

No install spec or external downloads; the code is bundled in the skill and will only be executed locally. This is the lower-risk install model.

Credentials

The script expects OPENAI_API_KEY (and optionally OPENAI_BASE_URL or OPENAI_API_BASE) but the skill metadata did not declare any required env vars or a primary credential. The API key requirement is appropriate for the task, but the omission in metadata is an incoherence and the optional API_BASE variables allow pointing to arbitrary endpoints — verify what base URL will be used before supplying secrets.

Persistence & Privilege

The skill does not request persistent/always-on inclusion and does not modify other skills or global configuration. It writes output files in the user's filesystem (~/Projects/tmp or ./tmp) which is expected for a generator.

What to consider before installing

This skill appears to be a straightforward OpenAI Images batch generator, but there are two things to check before running: (1) metadata vs runtime mismatch — the SKILL.md and script require OPENAI_API_KEY but the skill metadata does not declare that; ensure you only provide a key you trust and consider a scoped/restricted key. (2) the instructions use a hard-coded path (~/Projects/agent-scripts/...) which may run a different script on your machine — instead, run the bundled script included with the skill or inspect the script file contents first. To be safe: review scripts/gen.py yourself, run python3 scripts/gen.py with --dry-run to print prompts without making API calls, run in an isolated directory, and avoid supplying your primary API key until you confirm behavior. If you plan to use a custom API base, verify that OPENAI_BASE_URL/OPENAI_API_BASE are set to a trusted endpoint.

OpenAI Image Gen

Generate a handful of “random but structured” prompts and render them via OpenAI Images API.

Setup

• Needs env: OPENAI_API_KEY

Run

From any directory (outputs to ~/Projects/tmp/... when present; else ./tmp/...):

python3 ~/Projects/agent-scripts/skills/openai-image-gen/scripts/gen.py
open ~/Projects/tmp/openai-image-gen-*/index.html

Useful flags:

python3 ~/Projects/agent-scripts/skills/openai-image-gen/scripts/gen.py --count 16 --model gpt-image-1.5
python3 ~/Projects/agent-scripts/skills/openai-image-gen/scripts/gen.py --prompt "ultra-detailed studio photo of a lobster astronaut" --count 4
python3 ~/Projects/agent-scripts/skills/openai-image-gen/scripts/gen.py --size 1536x1024 --quality high --out-dir ./out/images

Output

• *.png images
• prompts.json (prompt ↔ file mapping)
• index.html (thumbnail gallery)