The Agent Skills Directory

[PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection because it fetches and processes content from the Wayback Machine, which archives arbitrary third-party websites. If an agent interprets the fetched text, it could be exposed to malicious instructions embedded in the archived pages.
Ingestion points: The fetch_archived_content function in scripts/search_archive.py retrieves content from the Internet Archive (web.archive.org), which reflects data from untrusted external sources.
Boundary markers: The script prints fetched content with simple summary headers but lacks explicit boundary markers or instructions advising the agent to ignore or isolate embedded instructions within the data.
Capability inventory: The script is limited to network reads and printing to standard output. It does not contain subprocess calls, file system write operations, or dynamic code execution capabilities across any of its files.
Sanitization: The extract_text function in scripts/search_archive.py performs basic sanitization by removing HTML tags and script/style blocks, which reduces the technical attack surface for code injection but does not prevent natural language prompt injection attacks.

web-archive-scraper