The Agent Skills Directory

[COMMAND_EXECUTION]: The skill utilizes several command-line utilities including git, gh, npm, pnpm, pytest, go test, and curl. These tools are used appropriately for their intended purposes: managing repository state, interacting with GitHub APIs, executing local test suites, and performing HTTP health checks on the deployed application.
[INDIRECT_PROMPT_INJECTION]: The skill has an attack surface for indirect prompt injection as it ingests untrusted data from external sources.
Ingestion points: PR metadata (titles, states) via gh pr view, commit history via git log, and response content from production URLs via curl.
Boundary markers: None explicitly defined in the instructions to delimit external data.
Capability inventory: The skill can perform file system operations (via test runners), merge code (gh pr merge), and push to remote repositories (git push).
Sanitization: No specific sanitization or filtering of the ingested strings is performed before they are used in reports or command arguments. However, the risk is significantly reduced by the disable-model-invocation: true configuration, which prevents the model from autonomously invoking tools based on malicious instructions found in the ingested data.

land-and-deploy