The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill processes user-supplied JSON files for timeline analysis, which are then interpolated into markdown reports. While the skill includes schema and type validation in modules/validators.py, it lacks explicit boundary markers or instructions to the agent to disregard potential instructions embedded within the input data. This creates a surface for indirect prompt injection if an attacker provides data containing malicious agent commands.
Ingestion points: project_timeline_calculator.py (line 64) and timeline_generator.py (line 82).
Boundary markers: No specific delimiters or "ignore instructions" warnings are used in the prompt or output formatting.
Capability inventory: File system write access for report generation (project_timeline_calculator.py line 203).
Sanitization: Manual schema validation and type checks are implemented in modules/validators.py.
[DYNAMIC_EXECUTION]: The skill uses sys.path.insert to include a local Shared_Utils directory located relative to the script's own path. This is a standard modularization pattern used to share code between different skills in a multi-skill environment. The paths are calculated deterministically from the file location and do not involve user-controlled input.

expropriation-timeline-expert