Agent Profile Images

AdvisoryAudited by Static analysis on Apr 30, 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.

Note

ASI04: Agentic Supply Chain Vulnerabilities

What this means

If applied blindly, source snapshots could overwrite or alter core OpenClaw gateway/UI behavior beyond what the user expected.

Why it was flagged

The skill is presented as source snapshots rather than a pinned installer or minimal patch, so applying it requires provenance and diff review.

Skill content

This skill packages the source-level feature implementation... The `references/` folder contains the feature implementation snapshots

Recommendation

Review the referenced files as diffs against your exact OpenClaw version and apply only the avatar-related changes you intend.

Note

ASI02: Tool Misuse and Exploitation

What this means

Avatar upload is expected, but weak validation in the implementation could allow oversized or invalid files to be submitted.

Why it was flagged

The upload RPC accepts user-supplied file metadata and image data; the visible schema does not itself bound MIME type, size, or image validity.

Skill content

filename: NonEmptyString,
contentType: NonEmptyString,
data: NonEmptyString

Recommendation

Ensure the backend implementation validates file type, size, decoding, and storage path before enabling the upload RPC.

Note

ASI03: Identity and Privilege Abuse

What this means

An operator with admin scope can change or remove stored agent avatars.

Why it was flagged

The avatar mutation methods are placed in the admin-only method group, meaning they require elevated operator authority.

Skill content

"agents.avatar.upload",
"agents.avatar.generate",
"agents.avatar.remove"

Recommendation

Keep admin gateway credentials limited to trusted users and confirm avatar changes are user-initiated.

Note

ASI07: Insecure Inter-Agent Communication

What this means

Custom avatar prompts or instructions may be processed by OpenAI according to that provider’s policies.

Why it was flagged

The feature uses an external image-generation provider, so generation themes/instructions may be sent outside the local OpenClaw environment.

Skill content

Adds `agents.avatar.generate` for themed image generation using OpenAI Images.

Recommendation

Avoid putting secrets or sensitive private details in avatar generation instructions, and verify which OpenAI credentials/account will be used.