The Agent Skills Directory

[PROMPT_INJECTION]: The skill exhibits 'context stuffing' across eight separate reference files (e.g., references/architecture-patterns.md, references/security-best-practices.md, references/state-management.md). Each file contains 150 repetitive sections of identical technical descriptions and code snippets.
Ingestion points: The skill ingests user preferences regarding database schemas and authentication providers in SKILL.md to generate configurations.
Boundary markers: Absent. The skill does not use delimiters to separate generated code or user-provided data from its own internal instructions.
Capability inventory: The skill is capable of generating shell commands for Docker and SQL DDL for database management, as seen in SKILL.md and references/hasura-postgrest.md.
Sanitization: Absent. User-provided preferences and variables are interpolated directly into the generated output templates without validation or escaping.
[COMMAND_EXECUTION]: The skill provides detailed instructions and templates for the agent to execute shell commands, specifically docker run and docker-compose, as well as SQL commands for database setup and policy configuration.
[EXTERNAL_DOWNLOADS]: The skill configuration templates reference several official images from Docker Hub, including hasura/graphql-engine, postgrest/postgrest, and postgres:16. These are reputable and well-known services used for their intended purpose.

data-data-api