retriever

This skill should be used when the user types "/bates", "/bates ABC0001-ABC0010", or "/bates clear". It exposes Retriever's visible slash-command surface for Bates-aware browsing.

This skill should be used when the user types "/columns", "/columns list", "/columns set ...", "/columns add ...", "/columns remove ...", or "/columns default". It exposes Retriever's visible slash-command surface for inspecting and changing displayed columns.

This skill should be used when the user types "/dataset", "/dataset list", "/dataset production", "/dataset production, priority", "/dataset clear", or "/dataset rename old-name new-name". It exposes Retriever's visible slash-command surface for dataset scoping, discovery, and renaming.

Use this skill when the user wants to manage the schema of custom document fields — list, add, rename, delete, re-describe, or change the type of a custom field — or when the user types "/field", "/field list", "/field add", "/field rename", "/field delete", "/field describe", or "/field type".

Use this skill when the user wants to populate, set, tag, mark, label, classify, annotate, flag, or clear a custom or editable built-in field value on one document or on a filtered/scoped result set — or when the user types "/fill".

This skill should be used when the user types "/filter", "/filter ...", or "/filter clear". It exposes Retriever's visible slash-command surface for SQL-like filter refinement.

This skill should be used when the user types "/from-run", "/from-run 42", or "/from-run clear". It exposes Retriever's visible slash-command surface for scoping results to a processing run.

Use this skill when the user wants Retriever to ingest a processed litigation or eDiscovery production such as a DAT/OPT/TEXT/IMAGES set, a Bates-numbered volume, or explicitly asks to run ingest-production.

Use this skill when the user wants to ingest, index, import, load, add, upload, or process files or folders into Retriever, or to refresh changed files, ingest a processed production, or explain what was ingested — including phrasings like "index this folder", "import these files", "load the Downloads directory", "add this PST to the collection", "upload and process this batch", "re-index the gmail-max mbox", or "what did you ingest". It bootstraps the workspace if needed, chooses between ingest and ingest-production, and summarizes the results.

This skill should be used when the user types "/next" or asks for the next page of the active Retriever browse results. It exposes Retriever's visible slash-command surface for forward pagination.

This skill should be used when the user types "/page-size" or "/page-size 25". It exposes Retriever's visible slash-command surface for inspecting and changing rows per page.

This skill should be used when the user types "/page", "/page 3", "/page first", "/page last", "/page next", or "/page previous". It exposes Retriever's visible slash-command surface for pagination.

Use this skill when Retriever needs to ingest, parse, normalize, or preview supported document types. It defines file-type support, encoding behavior, preview artifact rules, and per-file failure isolation.

This skill should be used when the user says "retriever ping", "test retriever", "is retriever installed", or "check retriever". Confirms the Retriever plugin is installed and responding.

This skill should be used when the user types "/previous" or asks for the previous page of the active Retriever browse results. It exposes Retriever's visible slash-command surface for backward pagination.

Use this skill when the user wants Retriever to ingest or inspect Outlook PST mail archives. It covers the first-class libpff-python / pypff-backed PST container pipeline, where one PST source expands into message documents plus attachment children.

Use this skill when the user wants to execute a planned Retriever processing run inline or in the background — including natural-language phrasings like "run OCR", "start the image-description job", "kick off the OCR worker", "process the queue", "execute run 42", "resume the stalled run", or "finish processing". It decides between inline execution and spawning a subagent, and drives the claim / complete / heartbeat loop against the processing DB.

Use this skill when the user wants to know the current Retriever tool or schema version for a workspace, or explicitly asks to run schema-version.

Use this skill when creating, inspecting, or migrating Retriever's SQLite schema. It is the source of truth for table layouts, path rules, custom field registry behavior, manual field locks, and schema-version behavior.

This skill should be used when the user types "/scope", "/scope list", "/scope clear", "/scope save ...", or "/scope load ...". It exposes Retriever's visible slash-command surface for inspecting and managing the active scope.

Use this skill when presenting Retriever search results or translating user intent into structured search filters. It defines paging, sorting, preview-link behavior, and the default result formats for browsing versus targeted lookups.

Use this skill whenever the user wants to show, list, view, display, browse, find, search, or otherwise retrieve documents, conversations, emails, chats, threads, messages, files, or attachments from the Retriever collection — including narrowing by keyword, phrase, sender, recipient, date, content type, dataset, Bates range, or any other field — or when the user types "/search", "/search query", "/search clear", "/search --within term", or "/search --fts token". It runs Retriever's search and slash browse commands with structured filters and returns the standard rendered table output.

This skill should be used when the user types "/sort", "/sort list", "/sort date_created desc", "/sort date_created desc, file_name asc", or "/sort default". It exposes Retriever's visible slash-command surface for sort inspection and changes.

Use this skill when working with Retriever's canonical tool bundle. It defines the pinned generated tools.py bundle, checksum expectations, and the current command surface.

Use this skill when initializing, checking, or refreshing a Retriever workspace. It defines the .retriever directory layout, bootstrap flow, runtime refresh rules, the pinned dependency contract for the current runtime, and the current workspace maintenance command surface.