text-extract-citations

HIGH W007: Insecure credential handling detected in skill instructions.

• Insecure credential handling detected (high risk: 0.90). The skill's examples explicitly show exporting a raw private key (FITTINGS_PRIVATE_KEY=0x...) and notes the key "sits in plaintext wherever the agent runs," which encourages embedding raw secrets in shell commands or outputs and can force an LLM to handle or emit the secret verbatim.

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

• Potentially malicious external URL detected (high risk: 0.80). The skill's README explicitly instructs downloading and then running remote code via "curl -sO https://fittings.sh/skill/call.mjs", which fetches executable JavaScript that would be run locally.

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

• Direct money access detected (high risk: 1.00). The skill includes a built-in payment flow that requires a crypto private key (FITTINGS_PRIVATE_KEY) and instructs the client to sign an authorization to spend USDC on Base; a facilitator then settles/broadcasts the transaction. This is an explicit crypto wallet/payment integration (off-chain signing + on-chain settlement), i.e. direct financial execution capability.