Social Accuracy Checker
Analysis
This instruction-only skill is coherent for fact-checking social drafts, with minor cautions around external web searches and a broader-than-needed edit permission.
Findings (2)
Artifact-based informational review of SKILL.md, metadata, install specs, static scan signals, and capability signals. ClawScan does not execute the skill or run runtime probes.
Checks for instructions or behavior that redirect the agent, misuse tools, execute unexpected code, cascade across systems, exploit user trust, or continue outside the intended task.
allowed-tools: - Read - Write - Edit - web_search - web_fetch
The skill is allowed to read and mutate files and use web tools. This mostly fits the purpose of reading drafts, checking facts online, and writing a report, but Edit is broader than the report-only instructions require.
Checks for exposed credentials, poisoned memory or context, unclear communication boundaries, or sensitive data that could leave the user's control.
Verify each claim via web search — confirm it's accurate, current, and not fabricated
The skill intentionally sends claim-derived queries to web search/fetch providers. That is appropriate for public-source fact-checking, but unpublished draft claims may be reflected in external search queries.
