The Agent Skills Directory

[SAFE]: No malicious patterns, prompt injections, or obfuscated content were detected in the skill instructions, metadata, or supporting documentation.
[COMMAND_EXECUTION]: The skill utilizes a local Python script, scripts/tax_package_summary.py, to aggregate the status of tax artifacts. The script is transparent, uses standard libraries (json, sys, collections, pathlib), and performs safe data aggregation without any dynamic execution or network activity.
[DATA_EXPOSURE]: The skill handles sensitive financial information (P&L, 1099 data, bank statements) as part of its primary function. All operations described are local to the agent's environment, and there are no instructions to exfiltrate this data or use it inappropriately.
[PROMPT_INJECTION]: The skill processes external financial data which represents a surface for indirect prompt injection.
Ingestion points: Financial records, receipts, and bank statements referenced in SKILL.md.
Boundary markers: None explicitly defined in the prompts.
Capability inventory: Includes file system access (Read/Write), WebFetch, and script execution.
Sanitization: The provided summary script uses standard JSON parsing, which prevents code-level injection, though the agent's processing of data remains a standard surface for this category of skill.

tax-prep