OpenClaw Key Management

Secure credential storage system for OpenClaw that encrypts and protects API keys, tokens, and sensitive credentials from memory file compromise.

Do not store production secrets with this skill yet. Before installing or trusting it: (1) Fix and verify the code — key_manager.sh references a non-existent key_vault_simple.js while the repo contains key_vault.js (the CLI will fail); (2) Confirm the master-key logic — the code always derives the master key from system identifiers and ignores the documented passphrase mode (so 'high-security' protection is not actually enforced); (3) Test install and migration in a disposable workspace — migration edits MEMORY.md automatically and will replace tokens (could corrupt or leak data if pattern matching is wrong); (4) Review and run the Node.js module manually to confirm encrypt/decrypt semantics and ensure vault files are saved and loaded correctly (there are mismatches between save/load encodings that will likely break vault persistence); (5) If you need real security, demand a patched version that: honors master_key_mode, removes hardcoded paths or makes paths configurable, stops referencing missing modules, and includes unit tests demonstrating correct load/save and migration behavior. If the author cannot explain or fix these issues, treat the skill as untrusted.

Type: OpenClaw Skill Name: openclaw-key-management Version: 1.0.3 The skill provides a credential encryption vault using AES-256-GCM, but its implementation contains significant security risks. The primary issue is in `scripts/key_manager.sh`, which generates temporary Node.js scripts by injecting unsanitized shell variables ($SECRET_NAME, $SECRET_VALUE) directly into JavaScript code blocks, making it highly vulnerable to code injection. Additionally, the scripts contain hardcoded workspace paths (e.g., `/zhaining`) and specific logic in `scripts/key_manager.sh` to target 'Instreet' API keys (`sk_inst_`), suggesting the tool is tailored for a specific environment or target. While no evidence of intentional data exfiltration was found, the poor handling of sensitive input in a security-focused tool is a major red flag.