The Agent Skills Directory

[COMMAND_EXECUTION]: The instructions direct the agent to perform standard development tasks using shell commands, including npm run build, npm run journey, grep, and git operations. These commands are integral to the skill's purpose of managing repository tasks.
[DYNAMIC_EXECUTION]: The playbooks/feature.md and playbooks/perf-issue.md files instruct the agent to delegate code implementation to subagents. This facilitates complex task decomposition and parallelization within the agent's execution environment.
[EXTERNAL_DOWNLOADS]: In playbooks/prototype.md, the skill permits the use of CDN-hosted dependencies when building throwaway prototypes. This behavior is scoped to an isolated scratch directory for experimental purposes and follows common development practices.
[INDIRECT_PROMPT_INJECTION]: The skill is designed to process external project documentation and feature specifications (e.g., docs/product/CONTEXT.md, features/F-<n>.md) which serves as a potential vector for indirect instructions.
Ingestion points: Feature contracts (features/F-<n>.md), context files (docs/product/CONTEXT.md), and verification artifacts (docs/verify/).
Boundary markers: The instructions do not specify the use of delimiters or specific warnings to ignore instructions embedded within these data sources.
Capability inventory: The agent possesses capabilities to execute shell commands, perform file system writes, commit to version control, and spawn subagents.
Sanitization: There is no mention of sanitization, filtering, or validation processes for the content ingested from these project files.

boring-mode