The Agent Skills Directory

[No Code] (SAFE): The skill consists entirely of Markdown documentation and process descriptions. No scripts (.py, .js), binaries, or configuration files are included.
[External Downloads] (SAFE): No external dependencies, package installation commands (npm, pip), or remote URLs were detected.
[Data Exposure] (SAFE): Although the skill describes handling sensitive financial data (bank statements, ledgers), it does not contain any code to exfiltrate data or hardcoded credentials.
[Indirect Prompt Injection] (LOW): The skill identifies a workflow involving the ingestion of external data (bank statements). While this is a potential attack surface for indirect prompt injection, the lack of implementation code makes the current risk negligible.
Ingestion points: Load bank statement and internal ledger (SKILL.md)
Boundary markers: None present
Capability inventory: No executable capabilities detected
Sanitization: Not applicable as no processing logic is provided

Accounts Reconciler