The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill's primary function is to read and execute tasks defined in docs/STORIES.md, creating an attack surface where malicious data could influence agent behavior.
Ingestion points: Reads user-provided story requirements and acceptance criteria from docs/STORIES.md to drive the autonomous loop.
Boundary markers: None identified. The instructions do not include delimiters or warnings to ignore potentially malicious instructions embedded within the story text.
Capability inventory: The skill has broad capabilities, including executing shell commands (git), running local PowerShell scripts (./build.ps1, tools/snap.ps1, tools/dev.ps1), modifying source code, and invoking other skills via the Skill tool.
Sanitization: No sanitization or validation of the content within docs/STORIES.md is performed before the agent acts upon it.
[COMMAND_EXECUTION]: The skill relies on executing local system commands and scripts to automate the build and commit process.
Executes project-specific PowerShell scripts: ./build.ps1, tools/snap.ps1, and tools/dev.ps1 for building, snapshotting, and environment management.
Directly executes Git CLI commands (git add -A, git commit) to persist changes to the repository.

improve-loop