The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The skill downloads the HyperExecute CLI executable directly from the vendor's official domain.
Evidence: SKILL.md and reference/hyperexecute.md provide instructions to fetch the binary from https://downloads.lambdatest.com/hyperexecute/linux/hyperexecute.
Note: This is a legitimate vendor resource used for the migration's target platform.
[COMMAND_EXECUTION]: The skill instructs the agent to execute multiple shell commands for repository scanning and build verification.
Evidence: Runs python3 scripts/detect.py <repo> to identify vendor fingerprints.
Evidence: Executes build tool commands such as mvn -q test-compile, npx playwright test --list, and dotnet build to verify the success of the migration.
Evidence: Grants execution permissions (chmod u+x) and runs the downloaded HyperExecute CLI.
[INDIRECT_PROMPT_INJECTION]: The skill processes untrusted data from the user's local repository, creating a surface for indirect prompt injection.
Ingestion points: scripts/detect.py reads and analyzes files throughout the provided repository path.
Boundary markers: No explicit boundary markers or instructions to ignore embedded commands are present in the detection or modification logic.
Capability inventory: The agent has capabilities to write to the file system, execute shell commands (including compilers and the migration CLI), and perform network operations.
Sanitization: Content from the scanned repository is not sanitized before being processed by the agent's logic, which could allow malicious instructions embedded in the source code to influence the migration workflow.

azure-playwright-migration-skill