The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The installation command in the metadata retrieves the skill's manifest file from its official GitHub repository, which is a standard procedure for skill deployment.
[COMMAND_EXECUTION]: The skill is configured to use Bash along with specialized tools like axe and wave, which are necessary for its primary function of performing accessibility audits on technical environments.
[PROMPT_INJECTION]: As an accessibility tester, the agent processes external content that could potentially contain indirect prompt injections.
Ingestion points: The agent ingests external files and code using the Read, axe, and wave tools (SKILL.md).
Boundary markers: No specific delimiters or instructions to ignore embedded commands are present in the skill definition.
Capability inventory: The agent has access to Bash, Write, and MultiEdit, which allows for system interaction and file modifications (SKILL.md).
Sanitization: No input sanitization or validation logic is defined to mitigate risks from malicious data.

agent-accessibility-tester