The Agent Skills Directory

[REMOTE_CODE_EXECUTION]: The skill uses npx to execute a tool named skills that downloads and installs additional agent capabilities from a third-party repository (samber/cc-skills-golang). This process involves fetching and integrating external code into the agent's operational environment.
[COMMAND_EXECUTION]: The skill generates a Justfile containing a shell script for the show_dockerignore target. This script performs automated system operations, including creating temporary directories (mktemp), setting process traps, and running docker build with a custom heredoc to inspect the local filesystem context.
[EXTERNAL_DOWNLOADS]: The skill recommends integrating immanuwell/dockerfile-roast into GitHub Actions, which involves downloading and executing a third-party linting tool during the CI/CD process.
[INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and process data from existing repositories, such as documentation, manifests, and source code. This untrusted data could contain instructions meant to subvert the agent's logic during project design.
Ingestion points: SKILL.md Section 1 ("Read repository instructions, status, existing docs, manifests, and source").
Boundary markers: The instructions lack explicit delimiters or "ignore embedded instructions" warnings for the ingested content.
Capability inventory: The skill has the ability to write to the filesystem, generate and execute shell commands via Justfile, and install external agent skills.
Sanitization: No sanitization or validation logic is defined for the ingested repository data.
[DYNAMIC_EXECUTION]: The skill dynamically generates repository configuration files, CI workflows, and Justfile scripts based on project specifications and user input provided at runtime.

golang-project-starter