dt-coaching-foundation
Pass
Audited by Gen Agent Trust Hub on Jul 2, 2026
Risk Level: SAFECOMMAND_EXECUTION
Full Analysis
- Command Execution: The skill provides instructions for the agent to execute local scripts (
Invoke-PptxPipeline.ps1andinvoke-pptx-pipeline.sh) to support the generation of PowerPoint decks. These commands are executed within the user's terminal and are tied to the skill's primary functionality. - System Configuration: A maintenance protocol is included to address potential PowerShell version incompatibilities. If an upgrade is required, the skill mandates an explicit user approval checkpoint before any changes are made to the shell environment.
- Indirect Prompt Injection Surface: The coaching workflow involves ingesting and processing user-provided research data and project artifacts. While this is a common pattern for AI agents, the skill minimizes potential risks by utilizing structured templates and requiring evidence-grounded synthesis.
- Data Isolation: The skill strictly scopes all persistent state files and generated artifacts to a project-specific directory (
.copilot-tracking/dt/{project-slug}/). This ensures that project data remains organized and does not interfere with other files in the workspace.
Audit Metadata