The Agent Skills Directory

[COMMAND_EXECUTION]: The skill executes local utility scripts located at ~/.claude/skills/nanostack/bin/ (specifically find-artifact.sh, find-solution.sh, and save-artifact.sh) to retrieve planning context and persist state.
[EXTERNAL_DOWNLOADS]: The skill provides documentation and instructions for using established development tools and libraries from well-known providers (e.g., Clerk, Vercel, Supabase, Stripe) via standard package managers like npx. These references are informative and target official services.
[PROMPT_INJECTION]: The skill processes potentially untrusted data from project files (e.g., package.json, .nanostack/stack.json) and outputs from previous agent steps to define the implementation scope.
Ingestion points: Project artifacts retrieved via find-artifact.sh, historical solutions from find-solution.sh, and local configuration files such as stack.json.
Boundary markers: There are no explicit instructions to use delimiters or 'ignore instructions' markers when reading data from external files or script outputs.
Capability inventory: The skill can execute local shell scripts and orchestrate the invocation of other automated skills (review, security, qa, ship) using the Skill tool.
Sanitization: The skill does not define specific validation or sanitization procedures for the content of the files it ingests.

nano