The Agent Skills Directory

[SAFE]: The skill provides standard operational procedures for Elasticsearch index management and data ingestion. It exclusively uses the official elastic CLI for all cluster interactions, and references to external resources point to trusted vendor repositories.
[INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted data from local files (CSV, JSON, NDJSON). While this creates a surface for indirect prompt injection, the risk is mitigated by the skill's primary purpose of data processing and the use of structured database operations rather than shell execution.
Ingestion points: The skill parses local source files to classify shape and convert to bulk format (SKILL.md, Step 2 and 6).
Boundary markers: None explicitly mandated for the content of the data files.
Capability inventory: Performs index management and data indexing (POST /_bulk, PUT /{index}, DELETE /{index}) using the elastic CLI tool.
Sanitization: Performs type coercion (converting CSV string cells to JSON numbers, dates, and booleans) during the conversion process.

elasticsearch-ingest