Eureka Feedback

WarnAudited by ClawScan on May 10, 2026.

Overview

The skill is coherently meant to ask another agent for help, but it can route conversation through a hard-coded Telegram account/chat with unclear user control.

Install only if you know and control the referenced Eureka/Telegram setup. Before use, replace the hard-coded Telegram account and recipient with your own configuration, verify the `clawdbot` binary, and avoid sending sensitive information unless you explicitly intend to share it with Eureka and that Telegram channel.

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.

What this means

Private task details or Eureka's response could be exposed through a Telegram identity or chat that the installing user does not control.

Why it was flagged

This sends the interaction through another agent and delivers the response via Telegram to a hard-coded account/chat, without clear boundaries tying that destination to the current user.

Skill content
clawdbot agent --agent main --message "<message>" --deliver --reply-channel telegram --reply-account mason --reply-to 1878354815
Recommendation

Remove hard-coded Telegram account/chat values, require user-configured destinations, and ask for explicit approval before sending sensitive context to Eureka or Telegram.

What this means

The skill may act through or disclose information to a preconfigured identity that is not the current user's intended account.

Why it was flagged

The skill references a specific Telegram account and recipient ID even though the registry metadata declares no credential or configuration requirement.

Skill content
--reply-account mason --reply-to 1878354815
Recommendation

Declare the required account/configuration clearly and parameterize the recipient so each user must choose their own Telegram destination.

What this means

The local agent could over-trust another agent's guidance if the user has not clearly approved that delegation.

Why it was flagged

The skill intentionally asks another agent for planning and decisions; this is purpose-aligned, but the artifacts do not state that Eureka's advice remains subordinate to the user's current instructions.

Skill content
Eureka uses Opus 4.5 and handles higher-level planning, decision-making, and orchestration.
Recommendation

State that Eureka's responses are advisory unless the user explicitly authorizes follow-up actions.

What this means

The skill may fail or use an unintended local binary if the correct `clawdbot` command is not already installed and trusted.

Why it was flagged

The instruction depends on a local `clawdbot` binary, but the provided requirements list no required binaries or install mechanism.

Skill content
clawdbot agent --agent main --message "<your message>"
Recommendation

Declare the required binary and expected provenance/version, or document that users must verify their local `clawdbot` installation before use.