The Agent Skills Directory

[SAFE]: The skill instructions and metadata do not contain any malicious patterns, obfuscation, or unauthorized access attempts. The core logic is limited to retrieving and filtering FreshBooks accounting data.
[INDIRECT_PROMPT_INJECTION]: The skill possesses a surface for indirect prompt injection as it ingests and displays data from the FreshBooks API, such as project titles and business names, which are externally controlled.
Ingestion points: Outputs from FRESHBOOKS_LIST_BUSINESSES and FRESHBOOKS_LIST_PROJECTS in SKILL.md.
Boundary markers: No explicit delimiters or instructions to ignore embedded content are present in the prompt templates.
Capability inventory: The skill does not include high-risk capabilities such as arbitrary file writing, shell command execution, or network exfiltration to untrusted domains.
Sanitization: There is no evidence of output sanitization for the data retrieved from the FreshBooks API.
[EXTERNAL_DOWNLOADS]: The skill requires a connection to an external MCP server hosted at https://rube.app/mcp. This is a functional requirement for the skill to bridge with the FreshBooks API and is documented as part of the setup process.

FreshBooks Automation