Skillv1.0.0

ClawScan security

News Sentiment Analysis · ClawHub's context-aware review of the artifact, metadata, and declared behavior.

BenignMar 16, 2026, 1:54 PM

Verdict: Benign
Confidence: high
Model: gpt-5-mini
Summary: This is an instruction-only skill that provides heuristics for judging A‑share news; it requests no credentials, has no install steps, and its requirements match its description.
Guidance: This skill is an offline, prose guide (no code) describing how to judge A‑share news and appears internally consistent. Before using it: (1) remember it’s a heuristic, not financial advice—test it on known examples; (2) decide and control where your agent will fetch market/flow data (trusted APIs only), since the SKILL.md doesn’t specify sources; (3) monitor any autonomous runs if the agent is allowed to act on its own, and do not provide private credentials (none are required by this skill). If a future version adds installs, env vars, or remote endpoints, re‑review for coherence and safety.

Purpose & Capability: okThe name/description and the SKILL.md all describe the same task (classifying news as 'real' vs 'fake' positive signals for A‑shares). There are no unrelated env vars, binaries, or install actions requested.
Instruction Scope: noteThe SKILL.md contains high‑level heuristics (e.g., check subsidy size relative to profit, review sector capital flows) but does not specify data sources or commands. It does not instruct reading local files, env vars, or sending data to external endpoints, but because it implies checking market/flow data an agent would need to fetch that data from somewhere—those data-source choices are not specified.
Install Mechanism: okNo install spec and no code files are present (instruction-only). Nothing will be written to disk or executed during install.
Credentials: okThe skill declares no required environment variables, credentials, or config paths; requested access is minimal and proportional to the stated purpose.
Persistence & Privilege: okalways is false and the skill does not request persistent or elevated platform privileges. It does not modify other skills or system configs.