The Agent Skills Directory

[PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection because it ingests and processes untrusted public data from external social media platforms.
Ingestion points: The skill uses x/tweets/search/recent and linkedin/search/posts (as specified in SKILL.md) to pull external content into the agent's context.
Boundary markers: While the skill organizes results into Markdown tables, it lacks explicit instructions or delimiters to warn the agent about potentially malicious instructions embedded within the social media posts.
Capability inventory: The skill's permissions are restricted to read-only search operations and do not include capabilities for file modification, system command execution, or unauthorized network writes.
Sanitization: The instructions do not define any sanitization, filtering, or validation processes for the ingested social media content before it is processed by the agent.
[EXTERNAL_DOWNLOADS]: The skill connects to the unifapi MCP server at mcp.unifapi.com and references installation through npx skills add unifapi-agent/agents. These resources are official components provided by the skill's author to facilitate data access.

buying-signal-monitor