The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The skill requires the installation of the agentcash package from the npm registry using npm install -g. Additionally, it uses npx agentcash@latest to dynamically fetch and execute the most recent version of the tool at runtime.
[COMMAND_EXECUTION]: The skill instructs the agent to execute shell commands to interact with the enrichment APIs and manage an internal wallet (e.g., npx agentcash@latest wallet info, npx agentcash@latest fetch).
[DATA_EXFILTRATION]: The skill is designed to transmit user-provided contact information, such as emails, names, and LinkedIn URLs, to external API endpoints at stableenrich.dev. While this is the intended functionality of the skill, it involves the transfer of PII to a third-party service.
[PROMPT_INJECTION]: The skill processes structured data from external enrichment APIs, which presents an indirect prompt injection surface.
Ingestion points: Data is ingested from various endpoints under https://stableenrich.dev/api/, including people-enrich and linkedin-scrape (found in SKILL.md).
Boundary markers: There are no clear boundary markers or instructions to the agent to treat the fetched data as untrusted or separate from its instruction set.
Capability inventory: The agent can execute CLI commands and file operations via the agentcash tool.
Sanitization: No sanitization or validation logic is specified for the data returned by the external APIs before the agent interprets it.

data-enrichment