The Agent Skills Directory

[DATA_EXFILTRATION]: The skill instructs the user to set a PINECONE_API_KEY environment variable. The provided Python scripts (upsert.py, quickstart_complete.py) read this variable to authenticate requests to Pinecone's cloud infrastructure, which is a standard procedure for this vendor's tools.
[COMMAND_EXECUTION]: The skill utilizes shell commands like export for configuration, cp for file management, and uv run to execute local Python scripts included in the skill package. These commands are transparent and align with the skill's purpose as a guided tour for developers.
[EXTERNAL_DOWNLOADS]: The Python scripts specify dependencies on pinecone and typer. These are reputable libraries that will be downloaded from official package registries during script execution, facilitating interaction with the Pinecone API.
[PROMPT_INJECTION]: The 'Assistant Quickstart' path processes user-provided documents (PDF, Markdown, Text, Docx) which presents a surface for indirect prompt injection. Adversarial content within these files could attempt to influence the agent's behavior during RAG operations.
Ingestion points: Step 1 of the Assistant Quickstart in SKILL.md allows the ingestion of external document files or the generation of sample documents into the agent context.
Boundary markers: Absent. The instructions do not define specific delimiters or instructions to the agent to disregard embedded directives within the documents.
Capability inventory: The skill possesses the capability to execute shell commands via Bash and perform file operations (Read/Write), as well as interact with the Pinecone API.
Sanitization: Absent. The skill does not explicitly perform validation or sanitization on the content of documents before they are uploaded to the Pinecone Assistant or queried.

pinecone:quickstart