keep-skill 0.8.1__tar.gz → 0.9.0__tar.gz

{keep_skill-0.8.1 → keep_skill-0.9.0}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: keep-skill
-Version: 0.8.1
+Version: 0.9.0
 Summary: Reflective memory - remember and search documents by meaning
 Project-URL: Homepage, https://github.com/hughpyle/keep
 Project-URL: Repository, https://github.com/hughpyle/keep
@@ -68,7 +68,7 @@ keep find "what's the rate limit?"
 
 # Track what you're working on
 keep now "Debugging auth flow"
-keep now -V 1                    # Previous context
+keep now -V 1                    # Previous intentions
 ```
 
 ---
@@ -131,9 +131,9 @@ keep get ID --history                  # All versions
 keep list --tag project=myapp          # Find by tag
 keep list --tags=                      # List all tag keys
 
-# Current context
+# Current intentions
 keep now                               # Show what you're working on
-keep now "Fixing login bug"            # Update context
+keep now "Fixing login bug"            # Update intentions
 ```
 
 ### Python API

{keep_skill-0.8.1 → keep_skill-0.9.0}/README.md

@@ -17,7 +17,7 @@ keep find "what's the rate limit?"
 
 # Track what you're working on
 keep now "Debugging auth flow"
-keep now -V 1                    # Previous context
+keep now -V 1                    # Previous intentions
 ```
 
 ---
@@ -80,9 +80,9 @@ keep get ID --history                  # All versions
 keep list --tag project=myapp          # Find by tag
 keep list --tags=                      # List all tag keys
 
-# Current context
+# Current intentions
 keep now                               # Show what you're working on
-keep now "Fixing login bug"            # Update context
+keep now "Fixing login bug"            # Update intentions
 ```
 
 ### Python API

{keep_skill-0.8.1 → keep_skill-0.9.0}/SKILL.md

@@ -1,6 +1,6 @@
 ---
 name: keep
-version: 0.8.1
+version: 0.9.0
 description: Reflective Memory
 homepage: https://github.com/hughpyle/keep
 runtime: python:3.12-slim
@@ -48,7 +48,7 @@ This work uses `keep` for memory and reflection, in order to become skillful.
 **Practice:** Before acting, check what you know. After acting, capture what you learned.
 
 \`\`\`bash
-keep now                     # Current context
+keep now                     # Current intentions
 keep find "topic"            # What do I know?
 keep update "learning"       # Capture insight
 \`\`\`
@@ -70,7 +70,7 @@ This skill provides memory. Use it — always, as practice.  Be skillful.
 
 Ground yourself:
 1. What kind of conversation is this? (Action? Possibility? Clarification?)
-2. Where am I? What's the current context?
+2. Where am I? What are my current intentions?
    ```bash
    keep now
    ```
@@ -126,7 +126,7 @@ This table is a summary. The full framework includes the Language-Action Perspec
 keep get "_system:conversations"
 ```
 
-The conversations document is foundational. It teaches you to recognize where you are in work, what stance to take, and how to notice when things shift. Internalize it.
+The 'conversations' document is foundational. It teaches you to recognize where you are in work, what stance to take, and how to notice when things shift. Internalize it.
 
 ### The Triple Check
 
@@ -146,7 +146,7 @@ keep update "Assumption was X, actually Y. Next time: Z" -t type=breakdown
 
 **Do this now — index the source teaching:**
 ```bash
-keep update "file://$PWD/docs/library/mn61.html" -t type=teaching -t topic=reflection --lazy
+keep update "file://$PWD/docs/library/mn61.html" -t type=teaching -t topic=reflection
 ```
 
 Then retrieve and read it:
@@ -178,7 +178,7 @@ You've read the core patterns (Layer 2). There's more foundational material.
 **Do this now — read and index:**
 ```bash
 keep get "_system:domains"
-keep update "file://$PWD/docs/library/true_person_no_rank.md" -t type=teaching -t topic=commentary --lazy
+keep update "file://$PWD/docs/library/true_person_no_rank.md" -t type=teaching -t topic=commentary
 keep get "file://$PWD/docs/library/true_person_no_rank.md"
 ```
 
@@ -194,21 +194,21 @@ The [docs/library/](docs/library/) directory contains additional seed material.
 
 ## Layer 4: Efficient Use
 
-### Track Your Current Context
+### Track Your Current Intentions
 
 Start each session by checking where you are:
 ```bash
 keep now
 ```
 
-Update it as your focus changes:
+Update it as your intentions change:
 ```bash
-keep now "Working on authentication bug in login flow"
+keep now "Improving the authentication flow; keep it simple"
 ```
 
 Check previous context if needed:
 ```bash
-keep now -V 1          # Previous context
+keep now -V 1          # Previous intentions
 keep now --history     # All versions
 ```
 
@@ -299,8 +299,8 @@ Don't dump everything into context. Navigate the tree:
 
 | Command | Purpose | Example |
 |---------|---------|---------|
-| `now` | Get/set current context | `keep now` or `keep now "status"` |
-| `now -V N` | Previous context versions | `keep now -V 1` or `keep now --history` |
+| `now` | Get/set current intentions | `keep now` or `keep now "intention"` |
+| `now -V N` | Previous intentions versions | `keep now -V 1` or `keep now --history` |
 | `find` | Semantic similarity search | `keep find "authentication flow" --limit 5` |
 | `find --id` | Find similar to existing item | `keep find --id "docid" --limit 3` |
 | `search` | Full-text search in summaries | `keep search "OAuth"` |
@@ -315,24 +315,7 @@ Don't dump everything into context. Navigate the tree:
 | `collections` | List all collections | `keep collections` |
 | `init` | Initialize or verify store | `keep init` |
 | `config` | Show configuration and store path | `keep config` |
-| `process-pending` | Process lazy summaries | `keep process-pending --all` |
-
-### Fast Indexing with `--lazy`
-
-When using local models (MLX), summarization is slow. Use `--lazy` for fast indexing:
-
-```bash
-keep update "file:///path/to/doc.md" --lazy
-keep update "insight" -t type=learning --lazy
-```
-
-The `--lazy` flag:
-- Stores immediately with truncated placeholder
-- Spawns background processor automatically
-- Full summary generated asynchronously
-- Search works immediately (embeddings are synchronous)
-
-**Use `--lazy` when:** indexing many items, using local MLX models, or when you don't need the summary right away.
+| `process-pending` | Process pending summaries | `keep process-pending --all` |
 
 ### Output
 

{keep_skill-0.8.1 → keep_skill-0.9.0}/keep/api.py

@@ -534,13 +534,17 @@ class Keeper:
         summary: Optional[str] = None,
         source_tags: Optional[dict[str, str]] = None,  # Deprecated alias
         collection: Optional[str] = None,
-        lazy: bool = False
     ) -> Item:
         """
         Insert or update a document in the store.
 
         Fetches the document, generates embeddings and summary, then stores it.
 
+        **Summary behavior:**
+        - If summary is provided, use it (skips auto-summarization)
+        - For large content, summarization is async (truncated placeholder
+          stored immediately, real summary generated in background)
+
         **Update behavior:**
         - Summary: Replaced with user-provided or newly generated summary
         - Tags: Merged - existing tags are preserved, new tags override
@@ -553,9 +557,6 @@ class Keeper:
             summary: User-provided summary (skips auto-summarization if given)
             source_tags: Deprecated alias for 'tags'
             collection: Target collection (uses default if None)
-            lazy: If True, use truncated placeholder summary and queue for
-                  background processing. Use `process_pending()` to generate
-                  real summaries later. Ignored if summary is provided.
 
         Returns:
             The stored Item with merged tags and new summary
@@ -615,17 +616,14 @@ class Keeper:
                 )
                 summary = summary[:max_len]
             final_summary = summary
-        elif lazy:
-            # Truncated placeholder for lazy mode
+        else:
+            # Large content: async summarization (truncated placeholder now, real summary later)
             if len(doc.content) > max_len:
                 final_summary = doc.content[:max_len] + "..."
+                # Queue for background processing
+                self._pending_queue.enqueue(id, coll, doc.content)
             else:
                 final_summary = doc.content
-            # Queue for background processing
-            self._pending_queue.enqueue(id, coll, doc.content)
-        else:
-            # Auto-generate summary
-            final_summary = self._get_summarization_provider().summarize(doc.content)
 
         # Build tags: existing → config → env → user (later wins on collision)
         merged_tags = {**existing_tags}
@@ -686,8 +684,8 @@ class Keeper:
                     tags=old_doc.tags,
                 )
 
-        # Spawn background processor if lazy (only if summary wasn't user-provided and content changed)
-        if lazy and summary is None and not content_unchanged:
+        # Spawn background processor if content was queued (large content, no user summary, content changed)
+        if summary is None and len(doc.content) > max_len and not content_unchanged:
             self._spawn_processor()
 
         # Return the stored item
@@ -703,7 +701,6 @@ class Keeper:
         tags: Optional[dict[str, str]] = None,
         source_tags: Optional[dict[str, str]] = None,  # Deprecated alias
         collection: Optional[str] = None,
-        lazy: bool = False
     ) -> Item:
         """
         Store inline content directly (without fetching from a URI).
@@ -713,7 +710,8 @@ class Keeper:
         **Smart summary behavior:**
         - If summary is provided, use it (skips auto-summarization)
         - If content is short (≤ max_summary_length), use content verbatim
-        - Otherwise, generate summary via summarization provider
+        - For large content, summarization is async (truncated placeholder
+          stored immediately, real summary generated in background)
 
         **Update behavior (when id already exists):**
         - Summary: Replaced with user-provided, content, or generated summary
@@ -728,9 +726,6 @@ class Keeper:
             tags: User-provided tags to merge with existing tags
             source_tags: Deprecated alias for 'tags'
             collection: Target collection (uses default if None)
-            lazy: If True and content is long, use truncated placeholder summary
-                  and queue for background processing. Ignored if content is
-                  short or summary is provided.
 
         Returns:
             The stored Item with merged tags and new summary
@@ -794,14 +789,11 @@ class Keeper:
         elif len(content) <= max_len:
             # Content is short enough - use verbatim (smart summary)
             final_summary = content
-        elif lazy:
-            # Content is long and lazy mode - truncated placeholder
+        else:
+            # Content is long - async summarization (truncated placeholder now, real summary later)
             final_summary = content[:max_len] + "..."
             # Queue for background processing
             self._pending_queue.enqueue(id, coll, content)
-        else:
-            # Content is long - generate summary
-            final_summary = self._get_summarization_provider().summarize(content)
 
         # Build tags: existing → config → env → user (later wins on collision)
         merged_tags = {**existing_tags}
@@ -860,8 +852,8 @@ class Keeper:
                     tags=old_doc.tags,
                 )
 
-        # Spawn background processor if lazy and content was queued (only if content changed)
-        if lazy and summary is None and len(content) > max_len and not content_unchanged:
+        # Spawn background processor if content was queued (large content, no user summary, content changed)
+        if summary is None and len(content) > max_len and not content_unchanged:
             self._spawn_processor()
 
         # Return the stored item
@@ -1366,14 +1358,14 @@ class Keeper:
 
     def get_now(self) -> Item:
         """
-        Get the current working context.
+        Get the current working intentions.
 
         A singleton document representing what you're currently working on.
         If it doesn't exist, creates one with default content and tags from
         the bundled system now.md file.
 
         Returns:
-            The current context Item (never None - auto-creates if missing)
+            The current intentions Item (never None - auto-creates if missing)
         """
         item = self.get(NOWDOC_ID)
         if item is None:
@@ -1394,13 +1386,13 @@ class Keeper:
         tags: Optional[dict[str, str]] = None,
     ) -> Item:
         """
-        Set the current working context.
+        Set the current working intentions.
 
-        Updates the singleton context with new content. Uses remember()
+        Updates the singleton intentions with new content. Uses remember()
         internally with the fixed NOWDOC_ID.
 
         Args:
-            content: New content for the current context
+            content: New content for the current intentions
             tags: Optional additional tags to apply
 
         Returns:

{keep_skill-0.8.1 → keep_skill-0.9.0}/keep/cli.py

@@ -251,7 +251,7 @@ def main_callback(
     )] = None,
 ):
     """Reflective memory with semantic search."""
-    # If no subcommand provided, show the current context (now)
+    # If no subcommand provided, show the current intentions (now)
     if ctx.invoked_subcommand is None:
         from .api import NOWDOC_ID
         kp = _get_keeper(None, "default")
@@ -683,10 +683,6 @@ def update(
         "--summary",
         help="User-provided summary (skips auto-summarization)"
     )] = None,
-    lazy: Annotated[bool, typer.Option(
-        "--lazy",
-        help="Fast mode: use truncated summary, queue for later processing"
-    )] = False,
 ):
     """
     Add or update a document in the store.
@@ -715,15 +711,15 @@ def update(
         parsed_tags = {**frontmatter_tags, **parsed_tags}  # CLI tags override
         # Use content-addressed ID for stdin text (enables versioning)
         doc_id = id or _text_content_id(content)
-        item = kp.remember(content, id=doc_id, summary=summary, tags=parsed_tags or None, lazy=lazy)
+        item = kp.remember(content, id=doc_id, summary=summary, tags=parsed_tags or None)
     elif source and _URI_SCHEME_PATTERN.match(source):
         # URI mode: fetch from URI (ID is the URI itself)
-        item = kp.update(source, tags=parsed_tags or None, summary=summary, lazy=lazy)
+        item = kp.update(source, tags=parsed_tags or None, summary=summary)
     elif source:
         # Text mode: inline content (no :// in source)
         # Use content-addressed ID for text (enables versioning)
         doc_id = id or _text_content_id(source)
-        item = kp.remember(source, id=doc_id, summary=summary, tags=parsed_tags or None, lazy=lazy)
+        item = kp.remember(source, id=doc_id, summary=summary, tags=parsed_tags or None)
     else:
         typer.echo("Error: Provide content, URI, or '-' for stdin", err=True)
         raise typer.Exit(1)
@@ -760,15 +756,15 @@ def now(
     )] = None,
 ):
     """
-    Get or set the current working context.
+    Get or set the current working intentions.
 
-    With no arguments, displays the current context.
+    With no arguments, displays the current intentions.
     With content, replaces it.
 
     \b
     Examples:
-        keep now                         # Show current context
-        keep now "What's important now"  # Update context
+        keep now                         # Show current intentions
+        keep now "What's important now"  # Update intentions
         keep now -f context.md           # Read content from file
         keep now --reset                 # Reset to default from system
         keep now -V 1                    # Previous version
@@ -891,7 +887,7 @@ def now(
         item = kp.set_now(new_content, tags=parsed_tags or None)
         typer.echo(_format_item(item, as_json=_get_json_output()))
     else:
-        # Get current context with version navigation and similar items
+        # Get current intentions with version navigation and similar items
         item = kp.get_now()
         version_nav = kp.get_version_nav(NOWDOC_ID, None, collection=collection)
         similar_items = kp.get_similar_for_display(NOWDOC_ID, limit=3, collection=collection)

{keep_skill-0.8.1 → keep_skill-0.9.0}/pyproject.toml

@@ -4,7 +4,7 @@ build-backend = "hatchling.build"
 
 [project]
 name = "keep-skill"
-version = "0.8.1"
+version = "0.9.0"
 description = "Reflective memory - remember and search documents by meaning"
 readme = "README.md"
 requires-python = ">=3.11,<3.14"