The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill is intended to be used when the agent is asked to perform an 'a11y audit' or 'improve accessibility' on web content, which constitutes untrusted external data.
Ingestion points: The agent ingests external web pages and source code for auditing (SKILL.md).
Boundary markers: The instructions do not specify the use of delimiters or 'ignore' instructions when processing the audited content.
Capability inventory: The agent typically uses browser tools or file-system access to perform these audits.
Sanitization: No specific sanitization or validation of the ingested content is defined to prevent instructions within the audited pages from influencing the agent's behavior.
[EXTERNAL_DOWNLOADS]: The skill suggests installing and using well-known accessibility auditing tools from a public registry.
Recommends the installation of '@axe-core/cli' via the npm registry (SKILL.md).
Recommends running the 'lighthouse' tool via npx (SKILL.md).
[COMMAND_EXECUTION]: The skill provides example commands for executing accessibility audits via the command line.
Example: npx lighthouse https://example.com --only-categories=accessibility in SKILL.md.
Example: axe https://example.com in SKILL.md.

accessibility