leveldb-ops

HIGH W007: Insecure credential handling detected in skill instructions.

• Insecure credential handling detected (high risk: 1.00). The skill explicitly instructs reading LevelDB LocalStorage/IndexedDB records and the example scripts/prints output record values (repr(v) / rec.value), which requires the agent to read and potentially emit secrets (tokens, cookies, passwords) verbatim.

MEDIUM W011: Third-party content exposure detected (indirect prompt injection risk).

• Third-party content exposure detected (high risk: 0.85). The required workflow reads outsider-authored free text from the user’s copied Chromium/Electron LevelDB store files (e.g., Local Storage/IndexedDB values) and passes the decoded strings/JSON into the agent’s LLM context via the scripts’ printed output (e.g., print(f" => {v!r}") / json.dumps(...)).

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

• Potentially malicious external URL detected (high risk: 0.90). The skill's setup instructs a runtime pip install that fetches and installs remote code from git+https://github.com/cclgroupltd/ccl_chrome_indexeddb.git (used by the scripts via "from ccl_chromium_reader import ..."), so external code would be executed as a required dependency.