The Agent Skills Directory

[Data Exposure] (LOW): The script scripts/generate_launch_plan.sh uses the whoami command to include the local system username in the generated launch plan. This is a minor disclosure of environment data intended for identifying the document owner.
[Indirect Prompt Injection] (SAFE): Scripts such as scripts/generate_launch_plan.sh and scripts/assess_launch_tier.sh ingest user input via read and interpolate it into files. While no sanitization is performed, the risk is minimal as the output is static markdown/text documentation. Ingestion points: read prompts in interactive shell. Boundary markers: None. Capability inventory: Local file creation and writing. Sanitization: None.
[Command Execution] (SAFE): The bash scripts utilize standard interactive shell features and file operations that are consistent with their stated purpose of project management.

technical-launch-planner