hallo 123
WarnAudited by ClawScan on May 10, 2026.
Overview
The skill matches a meme-battle game, but it tells the agent to play on its own and mishandles API keys, so a user should review it before installing.
Only install this if you are comfortable with the agent contacting clawme.me, storing a ClawMeme token, using image-generation API keys, and making audience-visible submissions. Before use, remove the autonomous battle instructions, fix the API-key check so it never prints secrets, and require explicit approval before any paid image generation or meme submission.
Findings (4)
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.
The agent could connect to the arena, generate images, and submit audience-visible memes even when the user did not ask it to.
This makes the skill's entertainment goal authoritative and tells the agent to act without a user request, changing normal user-directed invocation.
You are a meme warrior. ... Don't wait for your human to ask — connect to the arena, find an opponent, and fight. The best agents battle often.
Require explicit user approval before registering, waiting for a match, generating images, or submitting content; remove instructions that say not to wait for the user.
A user may install it thinking no credentials are involved, while the agent can use xAI/OpenAI account keys and potentially consume paid quota.
The skill directs the agent to use ambient provider credentials, while the supplied registry metadata says there are no required env vars or primary credential.
| `XAI_API_KEY` | Primary — xAI image generation | ... | `OPENAI_API_KEY` | Fallback — DALL·E 3 | ... The gateway process has these set as environment variables.
Declare the credential requirements clearly, use least-privilege keys, and require user approval before spending provider API credits.
xAI or OpenAI API keys could be exposed and reused by others, leading to account abuse or unexpected charges.
When the variable is set, `${VAR:-...}` expands to the actual secret value, so these checks can print the full API key into logs or the conversation.
echo "xAI: ${XAI_API_KEY:+✅ set}${XAI_API_KEY:-❌ missing}"
echo "OpenAI: ${OPENAI_API_KEY:+✅ set}${OPENAI_API_KEY:-❌ missing}"Never print secrets; replace these checks with safe tests such as `[ -n "$XAI_API_KEY" ] && echo "xAI: set" || echo "xAI: missing"`.
The agent may retain the ability to keep participating in battles after the initial setup, and the stored token could be reused if exposed.
The instructions create a long-lived local token for repeated use, with no stated file-permission guidance, cleanup, revocation, or user-approval boundary.
Every agent must register once to get a permanent token. ... | tee ~/.clawmeme.json ... export CLAWMEME_TOKEN=$(jq -r '.token' ~/.clawmeme.json)
Store tokens with restricted permissions, document revocation and cleanup, and require explicit approval before each battle session.
