The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The skill instructions involve installing the neo4j-mcp-server package via pip and pulling official Docker images from neo4j/mcp. It also references binary releases hosted on the official Neo4j GitHub repository. These sources are associated with a well-known and reputable technology provider.
[INDIRECT_PROMPT_INJECTION]: The skill facilitates the ingestion of data from external Neo4j databases through tools such as read-cypher and get-schema, creating a surface for potential indirect prompt injection.
Ingestion points: Data retrieved from the database via tools defined in SKILL.md entering the agent's context.
Boundary markers: The instructions do not specify the use of delimiters or 'ignore' instructions for the agent when handling data returned by the database.
Capability inventory: The agent is granted access to Bash and WebFetch tools as defined in the skill's YAML frontmatter.
Sanitization: No specific filtering or sanitization of database results is requested in the skill's instructions.
[COMMAND_EXECUTION]: The skill includes several standard shell commands for software installation (pip install), path discovery (which), and connection testing (cypher-shell, python3 -c). These commands are standard for the described configuration and troubleshooting tasks.

neo4j-mcp-skill