The Agent Skills Directory

[SAFE]: No security threats were detected across the skill's files. The core functionality is delivered through instructional markdown and local Python scripts.
[COMMAND_EXECUTION]: The skill instructions suggest running local Python scripts (tech_debt_analyzer.py and team_scaling_calculator.py). Analysis of these scripts confirms they are benign, performing only mathematical calculations and data processing using standard libraries (json, math, datetime). They do not perform network operations or unsafe file system writes.
[INDIRECT_PROMPT_INJECTION]: The skill is instructed to ingest external data from a company-context.md file if available. This represents a known attack surface where untrusted data could influence agent behavior. However, because the skill lacks dangerous capabilities (such as network egress or privilege escalation), the risk associated with this ingestion point is negligible. No explicit boundary markers were observed for this ingestion.
[DATA_EXPOSURE]: There are no hardcoded credentials, sensitive file paths, or network operations in any of the provided scripts or instructions.

cto-advisor