The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted external data in the form of database EXPLAIN plans (text or JSON) which could theoretically contain malicious instructions embedded in object names or metadata.
Ingestion points: SKILL.md Section 1 defines the ingestion of user-provided EXPLAIN text/JSON.
Boundary markers: Absent; there are no specific delimiters or warnings to ignore instructions embedded in the plans.
Capability inventory: The skill writes output files to docs/database-report/ and suggests SQL commands for execution.
Sanitization: No explicit sanitization or validation of the plan content is described.
[COMMAND_EXECUTION]: The skill generates SQL commands (e.g., CREATE INDEX) for performance tuning. While these are presented as recommendations for the user, they represent a capability to influence the host environment's database state.
[NO_CODE]: This skill consists entirely of markdown instructions and YAML configuration. It does not include any executable scripts (Python, JavaScript, shell, etc.), which significantly reduces the attack surface for categories like obfuscation or remote code execution.

query-explainer