valid-lei

HIGH W007: Insecure credential handling detected in skill instructions.

• Insecure credential handling detected (high risk: 0.80). The document shows and recommends exporting a private key into FITTINGS_PRIVATE_KEY (export FITTINGS_PRIVATE_KEY=0x...) and states the key "sits in plaintext wherever the agent runs", which encourages embedding a literal secret in shell commands or having the agent request/echo the secret—an explicit secret-handling pattern that risks exfiltration.

CRITICAL E005: Suspicious download URL detected in skill instructions.

• Suspicious download URL detected (high risk: 0.90). These URLs include a direct download of an executable script (call.mjs) from an unvetted personal service domain and the skill explicitly instructs you to curl and run it, which is a high-risk distribution pattern for malware.

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

• Potentially malicious external URL detected (high risk: 1.00). The SKILL.md instructs fetching and then running remote code via "curl -sO https://fittings.sh/skill/call.mjs", which downloads executable client code the skill may run, so this external URL is a runtime dependency that provides executable code.

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

• Direct money access detected (high risk: 1.00). The skill describes an explicit payment flow: the endpoint can return a 402 Payment Required with a signed quote that the client must sign, and the documentation shows setting FITTINGS_PRIVATE_KEY and running the client to pay (USDC on Base). It instructs using a wallet, holding funds, signing authorizations, and facilitator settlement — i.e., integration with crypto wallets/payment signing. This is a specific crypto payment capability, so it grants direct financial execution authority.