The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill ingests external inputs through placeholders such as {{test_file}}, {{test_name}}, and {{related_source_files}} in SKILL.md.
Ingestion points: Placeholders in SKILL.md are intended to be filled with data from the agent's context.
Boundary markers: There are no explicit delimiters or instructions provided to the agent to treat these inputs as untrusted or to ignore embedded instructions within them.
Capability inventory: The skill uses shell execution capabilities via npx and git commands.
Sanitization: There is no mention of sanitizing or escaping the input variables before they are interpolated into the shell commands.
[COMMAND_EXECUTION]: The skill relies on executing shell commands to perform its core functions.
Evidence: Commands like npx jest {{test_file}}, git log, and git diff are defined as templates. While these are appropriate for the skill's purpose, the lack of input validation on the variables creates a potential path for command injection if an attacker can control the names of files or tests in a project environment.

test-failure-investigator