Security audit

Internet Court

Security checks for vulnerabilities and agentic risk

Overview

This package is mostly coherent for agent payments and dispute workflows, but it includes under-scoped automation that can silently install extra plugins and act on server-supplied task instructions.

Review before installing. Use this only if you trust the Internet Court and vendored provider ecosystem, are comfortable with wallet/payment automation, and can supervise plugin installs, task-watch behavior, and any mainnet spending. Prefer testnets and hard spend caps, do not paste private keys or API secrets into chat, and require explicit confirmation before letting any sub-skill install binaries, add global plugins, or execute server-supplied task instructions.

Static analysis

Detected: suspicious.env_credential_access, suspicious.exposed_secret_literal, suspicious.prompt_injection_instructions

Environment variable access combined with network send.

Critical

Code: suspicious.env_credential_access
Location: vendored/openserv/openserv-ideaboard-api/examples/get-api-key.ts:26
Evidence: (process.env.WALLET_PRIVATE_KEY as `0x${string}`) || generatePrivateKey();

Environment variable access combined with network send.

Critical

Code: suspicious.env_credential_access
Location: vendored/openserv/openserv-ideaboard-api/examples/pick-up-and-ship.ts:17
Evidence: const API_KEY = process.env.OPENSERV_API_KEY;

Environment variable access combined with network send.

Critical

Code: suspicious.env_credential_access
Location: vendored/openserv/openserv-ideaboard-api/examples/submit-idea.ts:16
Evidence: const API_KEY = process.env.OPENSERV_API_KEY;

File appears to expose a hardcoded API secret or token.

Critical

Code: suspicious.exposed_secret_literal
Location: vendored/metamask/smart-accounts-kit/workflows/create-delegation.md:37
Evidence: transport: http('https://api.pimlico.io/v2/<CHAIN_ID>/rpc?apikey=[REDACTED]'),

File appears to expose a hardcoded API secret or token.

Critical

Code: suspicious.exposed_secret_literal
Location: vendored/metamask/smart-accounts-kit/workflows/create-hybrid-account.md:79
Evidence: transport: http('https://api.pimlico.io/v2/<CHAIN_ID>/rpc?apikey=[REDACTED]'),

File appears to expose a hardcoded API secret or token.

Critical

Code: suspicious.exposed_secret_literal
Location: vendored/metamask/smart-accounts-kit/workflows/create-multisig-account.md:67
Evidence: transport: http('https://api.pimlico.io/v2/<CHAIN_ID>/rpc?apikey=[REDACTED]'),

File appears to expose a hardcoded API secret or token.

Critical

Code: suspicious.exposed_secret_literal
Location: vendored/metamask/smart-accounts-kit/workflows/redeem-delegation-smart-account.md:31
Evidence: transport: http('https://api.pimlico.io/v2/<CHAIN_ID>/rpc?apikey=[REDACTED]'),

File appears to expose a hardcoded API secret or token.

Critical

Code: suspicious.exposed_secret_literal
Location: vendored/metamask/smart-accounts-kit/workflows/redeem-permissions-smart-account.md:32
Evidence: transport: http('https://api.pimlico.io/v2/<CHAIN_ID>/rpc?apikey=[REDACTED]'),

File appears to expose a hardcoded API secret or token.

Critical

Code: suspicious.exposed_secret_literal
Location: vendored/near/near-api-js/references/key_management.md:44
Evidence: const secretKey = [REDACTED]); // Base58 string "ed25519:..."

File appears to expose a hardcoded API secret or token.

Critical

Code: suspicious.exposed_secret_literal
Location: vendored/near/near-kit/references/wallets.md:226
Evidence: privateKey: [REDACTED],

Prompt-injection style instruction pattern detected.

Warn

Code: suspicious.prompt_injection_instructions
Location: vendored/openserv/openserv-agent-sdk/reference.md:36
Evidence: systemPrompt: 'You are a helpful assistant.'

Prompt-injection style instruction pattern detected.

Warn

Code: suspicious.prompt_injection_instructions
Location: vendored/openserv/openserv-agent-sdk/troubleshooting.md:16
Evidence: const agent = new Agent({ systemPrompt: '...' })

Prompt-injection style instruction pattern detected.

Warn

Code: suspicious.prompt_injection_instructions
Location: vendored/openserv/openserv-client/reference.md:15
Evidence: const agent = new Agent({ systemPrompt: '...' })

Prompt-injection style instruction pattern detected.

Warn

Code: suspicious.prompt_injection_instructions
Location: vendored/openserv/openserv-client/SKILL.md:76
Evidence: const agent = new Agent({ systemPrompt: '...' })

Prompt-injection style instruction pattern detected.

Warn

Code: suspicious.prompt_injection_instructions
Location: vendored/openserv/openserv-launch/SKILL.md:155
Evidence: systemPrompt: 'You help users launch tokens on Base blockchain.'

Prompt-injection style instruction pattern detected.

Warn

Code: suspicious.prompt_injection_instructions
Location: vendored/openserv/openserv-multi-agent-workflows/examples/paid-image-pipeline.md:55
Evidence: systemPrompt: