The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill provides instructions for handling database query results while mitigating injection risks.
Ingestion points: The skill processes database records and summaries returned via GraphDatabase.driver and execute_query in SKILL.md and references/async.md.
Boundary markers: None are explicitly defined for LLM context separation, but the documentation focuses on secure internal application logic and best practices.
Capability inventory: The skill is configured with Bash and WebFetch in the allowed-tools metadata in SKILL.md.
Sanitization: The skill mandates the use of parameterized queries ($param) to prevent Cypher injection and explicitly warns against using f-strings or other unsafe string formatting methods.
[CREDENTIALS_UNSAFE]: The skill demonstrates safe secret management by recommending the use of environment variables and providing guidance on using .env files with .gitignore to prevent accidental exposure of credentials.
[EXTERNAL_DOWNLOADS]: The skill refers to the installation of official and community-standard packages including the neo4j driver, the neo4j-rust-ext library for performance optimization, and python-dotenv for configuration management.

neo4j-driver-python-skill