The Agent Skills Directory

[SAFE]: The skill consists exclusively of markdown instructions and templates for product design. No executable scripts, binaries, or active command-line instructions were detected in the skill's workflow or reference files.
[EXTERNAL_DOWNLOADS]: The skill contains references and links to official documentation from well-known and trusted organizations, including W3C (WCAG standards), ISO (design standards), Microsoft (AI guidelines), Google (PAIR guidebook), and GOV.UK (design principles). These links are provided for informational context and do not involve automated downloads or runtime execution of remote code.
[DATA_EXPOSURE_AND_EXFILTRATION]: The skill includes specific security guardrails that prohibit the agent from requesting real credentials, payment information, or secrets. It explicitly mandates the use of synthetic fixtures (e.g., sample-user@example.test) for all design prototypes and usability study plans.
[INDIRECT_PROMPT_INJECTION]:
Ingestion points: The skill is designed to process user-provided evidence and product scope data within the templates/ and SKILL.md workflow.
Boundary markers: The skill instructs the agent to treat external request notes as untrusted data and provides clear entry checks to record missing inputs rather than inventing them.
Capability inventory: The skill has no capabilities to execute shell commands, perform network requests, or access sensitive system paths.
Sanitization: The methodology requires all design outputs to be traceable to specific evidence IDs, which acts as a logical check against the introduction of arbitrary or injected instructions.
[OBFUSCATION]: The evals/ directory contains SHA-256 hashes used for integrity verification of reference documents. These are standard cryptographic identifiers for version control and do not represent attempts to hide malicious content or URLs.

product-design-and-ux