Skillv1.0.0

ClawScan security

EODHD API · ClawHub's context-aware review of the artifact, metadata, and declared behavior.

Scanner verdict

SuspiciousMar 5, 2026, 3:12 PM

Verdict: suspicious
Confidence: medium
Model: gpt-5-mini
Summary: The skill mostly does what it claims (wraps the EODHD API) but has several inconsistencies and a token-leak risk that should be corrected before trusting it with your API credentials.
Guidance: This skill is plausibly what it says (an EODHD API wrapper) but has several issues you should address before installing or providing credentials: - Metadata mismatches: the registry metadata does not declare the required API token or the need to run Python; confirm you trust the source and that the environment has python3.11 and the 'requests' library. - Secret handling risk: the client writes the API token in plaintext to config.json and returns error dictionaries that include the 'params' field containing api_token. If the agent prints/logs or transmits error objects, your API key could be exposed. Consider: do not supply a long-lived token; prefer a limited-scope token or sandbox account; or modify the client to avoid returning or logging the token on error. - Operational suggestions: ask the publisher for provenance/homepage and for dependency/install instructions; edit the SKILL.md/metadata to declare the required credential (primaryEnv) or change the client to read the token from a declared env var instead of writing a file; add sanitization that removes api_token from any error or logging output. If you proceed anyway, only provide a limited/test API key, verify that the environment has the correct Python and dependencies, and review or modify the client to prevent token leakage in returned errors and logs.

Review Dimensions

Purpose & Capability: concernThe code and SKILL.md clearly require an EODHD API token and instruct writing it to config.json, but the registry metadata declares no required credentials or primaryEnv. The SKILL.md tells the agent to run python3.11, yet the skill metadata lists no required binaries. These metadata omissions make the declared requirements and the actual usage inconsistent.
Instruction Scope: noteRuntime instructions are focused on the stated purpose (calling the provided Python client) and ask the agent to write the user-provided API key into /home/ubuntu/skills/eodhd-api/config.json and to run Python scripts. The instructions do not request unrelated files or other credentials. However they rely on tool calls (file.write, shell executing python3.11) that are not reflected in metadata, and they give no guidance to avoid exposing secrets when printing or logging errors.
Install Mechanism: noteThere is no install spec (instruction-only install), which is low risk. However the included Python client depends on the 'requests' library and an explicit Python 3.11 invocation; no installation or dependency guidance is provided in SKILL.md or metadata, so runtime failures or surprise behavior are likely.
Credentials: concernThe skill requires the user's EODHD API token, but this is not declared in requires.env or primaryEnv. The token is written in plaintext to config.json inside the skill directory. More importantly, the client returns error objects containing 'url' and 'params' which include the api_token—this creates a realistic risk that the token will be exposed in logs, error messages, or outputs returned to the user or external systems.
Persistence & Privilege: okThe skill is not always-enabled and does not request elevated platform privileges. It writes its own config.json in its skill directory (normal). There is no evidence it modifies other skills or system-wide configs.