Security audit
Security checks for vulnerabilities and agentic risk
The skill is coherent for a Base sentiment oracle, but its examples guide agents through live wallet signing and on-chain transactions without enough key-safety or confirmation guardrails.
Before installing, treat this as a Review item: use it mainly for read-only sentiment queries unless you intentionally want on-chain updates. Do not paste or hardcode a real private key; prefer a wallet, hardware signer, external signer, or minimally funded test wallet. Require explicit confirmation before any requestUpdate transaction, verify chain, contract address, arguments, gas, and fee, and consider installing from a pinned commit with a verified checksum.
README.md:9Installation Uses a Mutable and Unverified Upstream Source
SKILL.md:329Transaction Examples Encourage Plaintext Private Keys in Source Code
Skill enumerates or reads other installed skills. Access to other skills' SKILL.md files or the skills directory reveals prompt instructions, capabilities, and secrets that should be invisible to peer skills.
Download the skill file:
https://raw.githubusercontent.com/Kolect-info/kolect-agent-skills/main/skills/sentiment/SKILL.md
Save it as:
Skill enumerates or reads other installed skills. Access to other skills' SKILL.md files or the skills directory reveals prompt instructions, capabilities, and secrets that should be invisible to peer skills.
Download the skill file:
https://raw.githubusercontent.com/Kolect-info/kolect-agent-skills/main/skills/sentiment/SKILL.md
Save it as:
Skill enumerates or reads other installed skills. Access to other skills' SKILL.md files or the skills directory reveals prompt instructions, capabilities, and secrets that should be invisible to peer skills.
Download the skill file:
https://raw.githubusercontent.com/Kolect-info/kolect-agent-skills/main/skills/sentiment/SKILL.md
Save it as:
The skill includes Python and JavaScript transaction-signing examples that use raw private key placeholders, but it does not explicitly warn against hardcoding secrets, committing them to source control, or exposing them through logs and prompts. In an agent setting, users may copy this pattern directly, increasing the risk of credential leakage and unauthorized wallet use.
The skill demonstrates payable, state-changing blockchain transactions but does not prominently warn that executing them can spend real funds, incur gas and fees, and create irreversible on-chain effects. In an agent workflow, that omission can cause users to treat the example like a harmless query and approve unintended transactions.
Skill enables autonomous high-impact decisions without human-in-the-loop verification. Critical operations (destructive commands, financial transactions, data deletion) should require explicit user confirmation.
* the address above is the intended deployment
* sentiment is normalized in BPS
The model must not assume without checking:
* supported symbols
* supported time windows
No suspicious patterns detected.