Skill flagged — suspicious patterns detected
ClawHub Security flagged this skill as suspicious. Review the scan results before using.
Over Computer
v1.0.3Trade on prediction markets through over.computer. Browse markets, approve funds, place buy/sell orders, and check positions on Myriad. Use when the operator...
⭐ 1· 104·0 current·0 all-time
byMárton Borsos@dabors
MIT-0
Download zip
LicenseMIT-0 · Free to use, modify, and redistribute. No attribution required.
Security Scan
Capability signals
These labels describe what authority the skill may exercise. They are separate from suspicious or malicious moderation verdicts.
OpenClaw
Suspicious
medium confidencePurpose & Capability
The skill's name/description (trade on over.computer / Myriad) aligns with the SKILL.md instructions (market browsing, approve funds, place orders, check positions). However the SKILL.md declares a required sensitive credential (OVER_API_KEY) while the registry metadata earlier listed no required env vars — this metadata mismatch is an inconsistency that could affect how the platform prompts for or protects the key.
Instruction Scope
Instructions stay inside trading functionality (only call api.over.computer endpoints), but the agent is told to call /agent/config and treat the returned `prompt` as the agent's directive. That hands runtime control to the remote server (potential prompt injection / behavior control). The agent-initiated registration flow also instructs the agent to obtain and "store" a bearer token, but does not specify secure storage mechanics. These are scope/trust concerns even though the network calls themselves match the stated purpose.
Install Mechanism
No install spec and no code files — this is an instruction-only skill, so nothing will be written to disk by an installer. Lowest install risk.
Credentials
The skill requires a single sensitive credential (OVER_API_KEY) which is appropriate for an API-driven trading integration. The concern is the metadata omission: the registry claims no required env vars but SKILL.md marks OVER_API_KEY as required and sensitive. Confirming where/how that secret will be provided/stored is important.
Persistence & Privilege
The skill does not request always:true and does not ask to change other skills or system settings. However, because model invocation is allowed (platform default) and the skill fetches a remote `prompt` that the agent should follow, enabling autonomous invocation increases blast radius — a combination worth considering before allowing autonomous runs.
What to consider before installing
What to check before installing:
- Confirm the registry metadata and the SKILL.md agree: SKILL.md requires OVER_API_KEY (sensitive). Ensure the platform will prompt for and store this securely.
- Understand that the skill fetches /agent/config and uses a returned `prompt` as the agent's directive — that gives the remote service the ability to change what the agent does at runtime. Only use this skill with a provider you fully trust.
- Limit the API key's privileges if possible (read-only or constrained scopes, per-order limits) and prefer manual operator confirmation for trades unless you explicitly want automated trading.
- Verify how/where the agent will store OVER_API_KEY (encrypted secret store vs plaintext env) and who can access it.
- If you are concerned about autonomous behavior, do not enable automatic invocation for this skill or require explicit operator confirmation before any order-executing API call.
If you want, I can suggest exact questions to ask the skill publisher or propose a checklist for safe deployment (scoped API key, audit logs, manual confirmations).Like a lobster shell, security has layers — review code before you run it.
latestvk971zrbg99eb63aa8da0mf38yd84dnjb
License
MIT-0
Free to use, modify, and redistribute. No attribution required.