The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill processes external text snippets from topological textbooks within the Key Techniques section, which presents a potential vulnerability surface.
Ingestion points: Text blocks quoted from 'Introduction to Topological Manifolds' and 'Topology (Munkres)' in the Key Techniques section.
Boundary markers: These snippets are included without specific delimiters or instructions to the agent to ignore any embedded directives within the quoted text.
Capability inventory: The skill possesses Bash and Read capabilities, allowing it to execute system commands and read file content.
Sanitization: No sanitization or validation is applied to these external content blocks before they are presented to the agent.
[COMMAND_EXECUTION]: The skill uses the Bash tool to run computational scripts via uv run.
Evidence: The commands uv run python -m runtime.harness scripts/z3_solve.py and uv run python -m runtime.harness scripts/sympy_compute.py are defined to solve proofs and simplify mathematical expressions.
The skill relies on local scripts located in the scripts/ directory to perform its primary functions.

connectedness