The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The skill instructs the agent to install the @cypress/cloud package using npm install -g. This is a vendor-owned resource originating from the skill's author, cypress-io, and is a standard requirement for the skill's stated purpose.
[COMMAND_EXECUTION]: The skill's primary function is executed through shell commands using the cy-cloud CLI. The instructions guide the agent to perform discovery and triage by passing user-supplied identifiers (such as projectId, runNumber, and testId) into these commands. This is the intended behavior of the skill.
[INDIRECT_PROMPT_INJECTION]: The skill processes untrusted data through the following evidence chain:
Ingestion points: User-provided URLs via --testResultUrl in investigation.md and test metadata retrieved from the Cypress Cloud API.
Boundary markers: The instructions do not define specific delimiters for separating user-provided identifiers from the rest of the prompt context.
Capability inventory: The agent can execute shell commands, read/write local files (screenshots/cache), and perform network requests via the CLI.
Sanitization: The skill suggests using double quotes around URL inputs in shell commands (e.g., cy-cloud test get --testResultUrl "<url>") to mitigate basic injection risks.
[CREDENTIALS_UNSAFE]: The skill manages Cypress Cloud authentication using Personal Access Tokens (PATs) and environment variables. It includes proactive security instructions, explicitly forbidding the agent from printing, echoing, or reading back tokens, and preventing the inspection of local credential files like ~/.config/cy-cloud/auth.json.

cypress-cloud-cli