The Agent Skills Directory

[PROMPT_INJECTION]: The skill processes user-provided issue descriptions via the $ARGUMENTS variable, which is then used to generate Git branch names and pull request metadata.
Ingestion points: The skill accepts external input via $ARGUMENTS in SKILL.md.
Boundary markers: The instructions lack explicit delimiters or instructions for the agent to ignore potentially malicious content embedded within the user-provided issue description.
Capability inventory: The agent is authorized to use gh and git commands and can execute a local smoke-test script within the cloned repository environment.
Sanitization: While branch names undergo basic character transformation, there is no comprehensive sanitization of the $ARGUMENTS input before it is interpolated into shell command strings for PR creation.- [COMMAND_EXECUTION]: The skill directs the agent to perform standard development tasks using the GitHub CLI (gh) and Git, such as cloning, branching, and creating pull requests for the Nairon-AI/flux-opencode repository.- [REMOTE_CODE_EXECUTION]: The workflow involves cloning the Nairon-AI/flux-opencode repository and executing a local script (./scripts/smoke-test.sh) to verify fixes. This is a standard verification step within the context of the vendor's own repository.

flux-opencode-contribute