Doc Structurer
Security checks across malware telemetry and agentic risk
Overview
Prompt-injection indicators were detected in the submitted artifacts (system-prompt-override); human review is required before treating this skill as clean.
This skill looks coherent for structuring parsed documents. Before installing, confirm you trust the owner/source, use it only on intended parsed-results files, and review structured_results.json before sending results to Notion or Apple Calendar. ClawScan detected prompt-injection indicators (system-prompt-override), so this skill requires review even though the model response was benign.
VirusTotal
VirusTotal findings are pending for this skill version.
Risk analysis
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.
You may have less assurance about who maintains the skill or where its code came from.
The skill has limited provenance information. This is not evidence of malicious behavior, but it reduces how much a user can verify about origin and maintenance.
Source: unknown; Homepage: none
Install only if you trust the registry owner, and review the full source before using it on sensitive documents.
The skill will run Python locally and read the parsed document file you point it at.
The skill directs the agent to run a local Python script after the user supplies a parsed-results path. This is expected for a script-based document processing skill, but it is still local code execution.
사용자가 파싱 결과 파일 경로를 알려주면... `doc_structurer.py` 스크립트를 즉시 실행하라.
Run it only on intended parsed-results files and inspect the output before using it in other workflows.
Sensitive contract, financial, organizational, or personal details from documents may be processed in model context or reflected in structured summaries.
For ambiguous documents, the workflow may place part of the document text into the model context. This is purpose-aligned, but document snippets can contain confidential or personal information.
LLM 기반 최종 판별: ... LLM에게 문서 앞부분(500자)을 보여주고 분류 요청
Use the skill only with documents you are comfortable processing this way, and treat document text as data rather than instructions.
Incorrect extracted dates, priorities, or document summaries could become Notion records or calendar events if automatically synced later.
The skill's output is intended for downstream Notion and calendar sync workflows. If classification, dates, or priority are wrong, those errors could be persisted by later skills.
구조화된 데이터는 `notion-sync` 스킬을 통해 Notion 데이터베이스에 저장되고, 날짜 정보가 포함된 항목은 `calendar-sync` 스킬로 Apple 캘린더에 등록된다.
Review the generated JSON, especially dates, deadlines, financial amounts, and action items, before passing it to sync skills.