The Agent Skills Directory

[COMMAND_EXECUTION]: The skill executes local system commands including docker compose, git, and PowerShell scripts to orchestrate the lifecycle of containerized stacks.
[DATA_EXFILTRATION]: The skill manages access to sensitive local environment files (.env) and host-based secret directories (UserSecrets). It provides these to the Docker daemon via the --env-file flag and volume mounts, which is a standard and secure practice for local container development to avoid hardcoding credentials. The documentation includes security warnings advising users to treat captured traffic (which may contain tokens) as sensitive.
[INDIRECT_PROMPT_INJECTION]: The skill processes external data that could theoretically be manipulated to influence its behavior.
Ingestion points: The skill reads the current git branch name using git rev-parse --abbrev-ref HEAD and directory structures to generate identifiers.
Boundary markers: There are no explicit markers used in the interpolation logic.
Capability inventory: The skill possesses the ability to execute PowerShell scripts, Docker commands, and write configuration files to the local file system.
Sanitization: The New-FsiStackOverride.ps1 script performs robust sanitization by using a regular expression to replace all non-alphanumeric characters in the branch name with hyphens before using it as a tag, which mitigates injection risks from maliciously named branches.

fsi-parallel-stack