The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The documentation URL https://jeffallan.github.io/claude-skills/skills/infrastructure/database-optimizer/ referenced in the skill is blacklisted by automated reputation scanners.
[COMMAND_EXECUTION]: Automated security scans identified SKILL.md as having a malicious file reputation (FileRepMalware). The skill provides instructions for the agent to recommend and potentially execute high-privilege database commands such as ALTER SYSTEM SET, which modifies system-level configurations.
[METADATA_POISONING]: The skill's frontmatter identifies the author as Jeffallan and links to an external GitHub profile, which contradicts the provided author context of harshsinghmp. Such discrepancies in authorship metadata can be used to mask the true origin of a skill.
[PRIVILEGE_ESCALATION]: The skill provides detailed instructions for performing operations that require administrative or superuser database privileges, including modifying system parameters and creating concurrent indexes, without sufficient warning about the security implications of these actions.
[INDIRECT_PROMPT_INJECTION]: The skill creates a high-risk surface for indirect prompt injection by design.
Ingestion points: The skill is designed to ingest and analyze potentially untrusted data from the user in the form of SQL queries and EXPLAIN ANALYZE execution plans.
Boundary markers: There are no explicit instructions or delimiters provided to the agent to distinguish between data being analyzed and instructions embedded within that data (e.g., within SQL comments).
Capability inventory: The skill possesses high-impact capabilities, including the generation and suggestion of system-level DDL/DML SQL commands and configuration changes.
Sanitization: The skill lacks any defined procedures for sanitizing or validating user-provided database objects or query text before processing.

database-optimizer