The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted data from the internet during the market research phase, creating a potential surface for indirect prompt injection.\n- Ingestion points: External market data retrieved via browser tools by the agents defined in trend-analyst.md, consumer-researcher.md, industry-analyst.md, and opportunity-finder.md.\n- Boundary markers: Absent; the instructions in SKILL.md and the agent files do not include delimiters or specific instructions to ignore embedded commands in the research data.\n- Capability inventory: The skill possesses the capability to write files to the local system and execute shell commands (specifically in the PDF generation workflow).\n- Sanitization: Absent; there is no explicit sanitization or validation of the retrieved web content before it is processed and synthesized into the final JSON report.\n- [COMMAND_EXECUTION]: The skill invokes a shell command to generate a PDF report using a script from a peer utility skill.\n- Evidence: SKILL.md instructs the agent to run python3 ${CLAUDE_PLUGIN_ROOT}/skills/media-utils/scripts/report_to_pdf.py using the analysis output as input.

market-researcher-agent