Skillv1.0.0

ClawScan security

Daily Geopolitical TikTok Reporter · ClawHub's context-aware review of the artifact, metadata, and declared behavior.

Scanner verdict

BenignApr 24, 2026, 10:50 PM

Verdict: benign
Confidence: high
Model: gpt-5-mini
Summary: The skill's code, runtime instructions, and requirements are coherent with its stated purpose of fetching free RSS news and generating TikTok scripts; there are only minor documentation/configuration inconsistencies and no signs of credential requests, hidden endpoints, or unexpected persistence.
Guidance: This skill appears to do what it says: fetch public RSS feeds and generate TikTok-ready scripts. Before installing or running it: 1) Run in a virtual environment (python venv) and inspect requirements.txt; avoid using --break-system-packages. 2) Review and, if needed, correct the RSS URLs and region/topic lists in news_gathering/fetch_news.py and config/config.yaml (SKILL.md mentions some files that are not present). 3) Be aware the script performs network requests to public RSS feeds and writes files to the working directory (generated_scripts_*). 4) If you plan to schedule it (cron/systemd/heartbeat), keep it sandboxed and rate-limit requests to avoid accidental scraping overload. 5) Review generated content for factual accuracy and legal/copyright concerns before publishing (geopolitical content can be sensitive). If you want higher assurance, run the code in an isolated environment and scan it with your own tools; if the skill later asks for credentials or includes remote downloads, treat that as a new risk and re-evaluate.

Review Dimensions

Purpose & Capability: noteThe skill's name/description (daily geopolitical TikTok scripts) matches the code: it fetches free RSS feeds, scores relevance/importance, and generates short scripts. Minor inconsistencies: SKILL.md/README reference templates (templates/*.yaml) and config files (regions.yaml, topics.yaml, schedule.yaml) and reference files under references/ that are not present in the package; config/config.yaml exists instead. These are documentation/config mismatches but not evidence of malicious intent.
Instruction Scope: okRuntime instructions and SKILL.md stay within the stated purpose (gather news, generate and save scripts, optionally schedule). The included code fetches public RSS feeds, processes them, and writes output files to the current working directory. There are no instructions to read unrelated system files or to access environment variables or credentials.
Install Mechanism: okThis is an instruction-and-code skill with no install spec in the registry. Dependencies are standard Python packages listed in requirements.txt (feedparser, pyyaml, python-dateutil, requests). No remote archives, URL-shorteners, or arbitrary installers are used in the package itself.
Credentials: okThe skill declares no required environment variables, credentials, or special config paths. The code operates without secrets. It does perform network requests to public RSS endpoints (list embedded in code), which is proportional to its purpose.
Persistence & Privilege: okThe skill does not request 'always: true' and is user-invocable. It writes output files to a local generated_scripts_* directory but does not modify other skills, system-wide agent settings, or install background services by itself. README suggests optional cron/systemd scheduling which would be user-controlled.