The Agent Skills Directory

[COMMAND_EXECUTION]: The workflow file workflows/schedule.md contains bash command templates using curl, grep, and cut. These are intended for the agent to execute when retrieving API keys from .env files and interacting with the Typefully API.
[DATA_EXFILTRATION]: The skill is designed to send user-generated content and metadata to api.typefully.com. Although this is the skill's primary function, it involves transmitting potentially sensitive content to an external service domain that is not on the standard whitelist.
[PROMPT_INJECTION]: The skill allows for the ingestion of external data via @file references in the workflows/essay.md workflow, which creates a potential surface for indirect prompt injection attacks.
Ingestion points: External files referenced by the user or extracted from the filesystem in the workflows/essay.md workflow.
Boundary markers: The skill lacks explicit delimiters or instructions for the agent to ignore any embedded instructions or malicious prompts within the ingested files.
Capability inventory: The skill possesses the ability to execute network requests via curl, write files to the ~/CybosVault/ directory, and utilize MCP tools for web searching and image generation.
Sanitization: There is no evidence of sanitization, validation, or escaping logic applied to the content retrieved from external file references before it is processed by the agent.

content