tempo

HIGH W007: Insecure credential handling detected in skill instructions.

• Insecure credential handling detected (high risk: 1.00). The skill explicitly instructs the agent to relay auth URLs, verification codes, and fund URLs (ephemeral secrets/tokens) back to the user verbatim, requiring the LLM to output sensitive credential-like values.

CRITICAL E005: Suspicious download URL detected in skill instructions.

• Suspicious download URL detected (high risk: 0.80). The list contains multiple high-risk entries: domains named "evil"/"malicious" and URLs with embedded credentials or non-printable control characters (obfuscation), plus direct executable downloads from GitHub releases (and curl|bash install patterns referenced in the repo) — all are common vectors for distributing malware or credential leakage.

MEDIUM W012: Unverifiable external dependency detected (runtime URL that controls agent).

• Potentially malicious external URL detected (high risk: 0.90). SKILL.md instructs agents/users at runtime to fetch and execute remote content—specifically curl -fsSL https://tempo.xyz/install | bash (executes remote code) and to run curl -fsSL https://tempo.xyz/SKILL.md to retrieve exact setup instructions that control the agent workflow—so these URLs are runtime external dependencies that can control prompts or execute code.

MEDIUM W009: Direct money access capability detected (payment gateways, crypto, banking).

• Direct money access detected (high risk: 1.00). The skill explicitly exposes wallet and payment functionality: a Tempo wallet with balances and funding commands, automatic payment handling, wallet-backed virtual cards, Stripe/bridge card configuration and issuing, card transactions/authorizations, and an approvals/allowance flow. These are specific financial APIs/commands to move funds, create/manage virtual cards, and perform transactions (not merely generic HTTP or browser automation). Therefore it grants direct financial execution authority.

MEDIUM W013: Attempt to modify system services in skill instructions.

• Attempt to modify system services in skill instructions detected (high risk: 0.80). The skill instructs an agent to run a remote installer via "curl ... | bash" and to perform local CLI login that opens a browser on the host—actions that modify the host environment and can change system state or request elevated privileges.