Link Midjourney Instagram
AdvisoryAudited by Static analysis on May 10, 2026.
Overview
No suspicious patterns detected.
Findings (0)
Artifact-based informational review of SKILL.md, metadata, install specs, static scan signals, and capability signals. ClawScan does not execute the skill or run runtime probes.
An agent run could publish images and captions to the user's Instagram account before the user separately reviews the exact outputs.
The skill directs the agent to execute the full automation, and the documented behavior includes public Instagram posts without an artifact-backed confirmation or review guard.
then post each PNG as its own Instagram web post ... 執行 pipeline（給 agent 用 `exec`） ... `python main.py`
Use `--skip-instagram` for dry runs, require explicit approval of generated images and captions before posting, and document a confirmation step.
If the repository code or local credential files are mishandled, the user's Instagram or Midjourney accounts could be accessed or used.
The skill requires direct Instagram credentials and stores reusable Midjourney/Instagram browser sessions, granting the automation durable account authority.
`.env` 內需 `INSTAGRAM_USERNAME`、`INSTAGRAM_PASSWORD` ... session 會存到 `browser_data/midjourney/` ... `browser_data/instagram/`
Use dedicated low-risk accounts where possible, avoid storing main account passwords, protect and rotate credentials, and inspect the runnable code before adding secrets.
Users must trust external or local code and packages not assessed in this skill review before granting account credentials and posting authority.
The skill tells users or agents to install dependencies and run repository code, but the supplied artifact set contains only SKILL.md, so the code and dependencies that handle credentials and posting are not reviewable here.
repository root ... 含有 `main.py` ... `pip install -r requirements.txt` ... `python main.py`
Provide the runnable code and dependency lockfiles for review, pin dependencies, document provenance, and verify `main.py` and `requirements.txt` before use.
Generated images or post content may leave the local/browser environment for third-party captioning.
If vision captioning is enabled by existing provider keys, generated images may be sent to Gemini or OpenAI for captioning; this is disclosed and purpose-aligned but should be understood.
若設了 `GEMINI_API_KEY`/`GOOGLE_API_KEY` 且未明確關閉，預設會開 vision ... `VISION_CAPTION_PROVIDER=openai` 搭配 `OPENAI_API_KEY`
Disable vision captioning or remove provider keys if you do not want images sent to an external vision API, and review the provider's data policy.