The Agent Skills Directory

[SAFE]: No malicious patterns or security vulnerabilities were detected during the analysis.
The prompt instructions are aligned with the intended project management persona and contain no attempts to bypass safety filters or override system constraints.
The included Python scripts (project_health_dashboard.py, risk_matrix_analyzer.py, resource_capacity_planner.py, stakeholder_mapper.py) utilize standard libraries and perform purely computational tasks without dangerous functions like eval(), exec(), or subprocess calls.
No hardcoded credentials, sensitive file access (e.g., SSH keys, AWS config), or network exfiltration patterns were found.
All external resource references in the documentation point to internal skill paths or legitimate project management methodologies.
[INDIRECT_PROMPT_INJECTION]: The skill ingests project-specific data from JSON files, which is a common surface for indirect prompt injection if the data originates from untrusted sources.
Ingestion points: Four Python scripts in the scripts/ directory ingest data from local JSON files via the data_file argument.
Boundary markers: While the scripts parse structured data, the skill does not provide explicit boundary markers or instructions to the agent to ignore potentially malicious text within the analyzed output.
Capability inventory: The skill is restricted to local data processing and text generation; it lacks high-risk capabilities such as network access, file-system modification, or shell execution.
Sanitization: The scripts include basic data validation, ensuring that numeric inputs for probability, impact, and health dimensions fall within expected ranges before processing.

senior-pm