The Agent Skills Directory

[COMMAND_EXECUTION]: The skill invokes python3 scripts/validate_design.py to perform structured validation of the DESIGN.md artifact. It also uses git log --oneline -20 to retrieve historical context regarding visual updates in the repository. These operations are local and consistent with the skill's auditing purpose.
[EXTERNAL_DOWNLOADS]: The skill references platform-specific scripts such as scripts/self_audit.py and scripts/generate_contracts.py. While these are not contained within the skill's own file list, they are presented as local dependencies within the expected agentera framework environment.
[PROMPT_INJECTION]: The skill exhibits an indirect prompt injection surface due to its requirement to ingest and analyze untrusted content from the project codebase.
Ingestion points: SKILL.md Step 1 reads the repository structure, component files, and vision documents (VISION.md, CLAUDE.md) to inform the design process.
Boundary markers: Absent; there are no specific instructions or delimiters used to prevent the agent from obeying instructions that might be embedded in the analyzed code files.
Capability inventory: The skill can execute local Python scripts and read/write project documentation.
Sanitization: No explicit sanitization or filtering of external content is performed before it is interpolated into the agent's context.

visualisera