engram-ingestlisted

# engram-ingest Load large volumes of content into Engram memory without hand-crafting individual `memory_store` calls. Supports local files and directories, exported chat histories from Slack, Claude, and ChatGPT, and full memory exports for backup. These tools are hidden from the normal tools/list. This skill calls them directly via HTTP. ## When to Use - User says "ingest this file", "load this directory into memory", "import my notes" - User says "parse my Slack export", "import my Claude history", "ingest ChatGPT export" - User says "export all memories", "back up Engram", "dump memories to files" - User says "check ingest status", "how is the ingestion job doing", "is the import done" - Any bulk load of existing documents or conversation history ## How to Use ### 1. Ingest a local file or directory ```bash xh POST "${ENGRAM_BASE_URL:-http://localhost:8788}/mcp" \ "Authorization: Bearer $ENGRAM_API_KEY" \ Content-Type:application/json \ jsonrpc=2.0 id:=1 method=tools/call \ params:='{"name":"memory_ingest","arguments":{"path":"<file-or-directory-path>","project":"<project>"}}' ``` `path` can be an absolute path to a single file or a directory. When a directory is given, Engram walks it recursively and ingests all supported files. The response includes a `job_id` for async status checks. **Supported file types**: Markdown, plain text, JSON, and other document formats Engram knows how to parse. ### 2. Ingest an exported chat history ```bash xh POST "${EN