The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill processes untrusted email content which serves as an entry point for malicious instructions. \n- Ingestion points: The scripts/agent.py file ingests data from external EML files provided via the --eml argument, specifically parsing headers, body text, and attachments. \n- Boundary markers: No explicit delimiters or instructions are provided to the agent to disregard instructions embedded within the analyzed email content. \n- Capability inventory: The skill possesses the ability to make network requests to external APIs (VirusTotal, urlscan.io) and write JSON reports to the local filesystem. \n- Sanitization: No sanitization or filtering is applied to the extracted email content before it is included in the incident report. \n- [METADATA_POISONING]: There is a discrepancy in the author information provided across the skill files. \n- Evidence: The SKILL.md file identifies the author as 'mahipal', whereas the LICENSE file attributes copyright to 'mukul975'. This inconsistency represents misleading metadata. \n- [EXTERNAL_DOWNLOADS]: The skill interacts with external security services to analyze suspicious indicators. \n- Evidence: scripts/agent.py uses the requests library to communicate with virustotal.com and urlscan.io. \n- Note: These are well-known security services, and their use is consistent with the skill's primary purpose of phishing incident response.

conducting-phishing-incident-response