The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill's primary function involves reading and analyzing the content of test files, which constitutes processing untrusted external data.
Ingestion points: Test files and configuration files are ingested via the Read tool and Grep commands as specified in the execution steps and agentic optimizations.
Boundary markers: The instructions do not specify the use of delimiters or 'ignore' instructions to prevent the agent from following malicious commands that might be embedded in the test files being analyzed.
Capability inventory: The skill has access to shell execution (limited to specific test runners) and filesystem write access via TodoWrite, which could be targeted by a successful injection.
Sanitization: There is no mention of sanitizing or filtering the content of the files before they are presented to the agent for analysis.
[DYNAMIC_CONTEXT_INJECTION]: The skill uses dynamic context execution to automatically populate information when the skill is loaded.
Evidence: The Context section in SKILL.md uses the ! syntax to run find commands that locate test files and configuration files in the repository.
Nature: The commands used (find . -type f ...) are benign discovery operations that do not incorporate user-provided arguments, posing no immediate security risk but documented for visibility.

code-test-quality