mneme-cli 0.5.0__tar.gz → 0.5.1__tar.gz

{mneme_cli-0.5.0 → mneme_cli-0.5.1}/AGENTS.md

@@ -213,6 +213,10 @@ out manually, then run `resync-resolve`.
 mneme tags suggest <client>/<page>                     # build tag packet
 mneme tags suggest <client>/<page> --json              # raw dict
 mneme tags apply <client>/<page> --add t1,t2 --remove t3
+
+# Bulk variants -- packet up to N pages in one round-trip
+mneme tags bulk-suggest --client <c> --filter req- --limit 50 --out packet.md
+mneme tags bulk-apply response.json     # response: {"pages": [{wiki_path, add, remove}, ...]}
 ```
 
 `mneme tags suggest` builds a **tag packet**: the page content, current
@@ -245,6 +249,37 @@ mneme tags list                                        # all tags + counts
 mneme tags merge <old> <new>                           # rename across all pages
 ```
 
+### 3.7 ENTITY — agent-driven classification
+
+```bash
+mneme entity suggest --client <c> --limit 50          # classification packet
+mneme entity apply --id iso-13485 --type standard     # one at a time
+mneme entity bulk-apply classifications.json          # batch
+```
+
+`mneme entity suggest` builds an **entity packet**: every `unknown`-typed
+entity, the workspace's current type distribution, the valid type
+vocabulary, an example wiki page per entity, and a prompt. The agent
+returns a JSON array of `{id, type}` objects which `bulk-apply` writes
+back atomically. Same philosophy as tags: mneme stays deterministic, the
+LLM does the classification.
+
+Valid types: `standard`, `company`, `person`, `product`, `technology`,
+`concept`, `brand`, `unknown`.
+
+### 3.8 HOME — generated landing page
+
+```bash
+mneme home --client <c>          # wiki/<c>/HOME.md
+mneme home --all-clients         # wiki/HOME.md (cross-client)
+```
+
+Generates an Obsidian-friendly navigation hub with Dataview queries
+(group by type, by ID prefix like `REQ-*` / `DDS-*`, top tags) and a
+plain-markdown `<details>` fallback so the page is useful outside
+Obsidian. Run after a large ingest, or whenever the wiki's shape
+changes meaningfully.
+
 ---
 
 ## 4. Profiles and the writing-style contract

{mneme_cli-0.5.0 → mneme_cli-0.5.1}/CHANGELOG.md

@@ -58,6 +58,39 @@ and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0
 - Chunking logic (`chunk_body`, `MAX_CHUNK_SIZE`, frame management).
 - Tantivy-reserved-word query sanitizer (FTS5 has different syntax).
 
+## [0.5.1] - 2026-04-14
+
+### Added
+- **`mneme entity suggest` / `entity apply` / `entity bulk-apply`** — agent-driven
+  entity classification (same packet pattern as `tags suggest`). Mneme builds a
+  packet of unclassified entities + the workspace type taxonomy + example pages;
+  the LLM agent classifies; mneme writes the types back atomically.
+- **`mneme tags bulk-suggest` / `tags bulk-apply`** — operate on many pages at
+  once. `bulk-suggest --client X --filter req- --limit 50` packets up to 50
+  matching pages; agent returns one JSON file; `bulk-apply` runs all the changes
+  with per-page error tolerance. Critical for tagging workspaces of hundreds of
+  pages.
+- **`mneme home --client <slug>` / `--all-clients`** — generates a `HOME.md`
+  navigation hub with Obsidian Dataview queries (group by type, by ID prefix
+  like REQ-*/DDS-*, top tags) plus a plain-markdown `<details>` fallback for
+  non-Obsidian viewers.
+- **`mneme ingest-dir --preserve-structure`** — mirrors source directory
+  hierarchy into wiki subdirectories. `sources/client/REQUIREMENTS/req-001.md`
+  becomes `wiki/client/requirements/req-001.md` instead of flattening. Also
+  resolves same-basename-different-directory collisions naturally.
+- **`mneme resync` auto-detects subpath** from a source's location under
+  `sources/<client>/`, so resyncs of preserve-structure ingests target the
+  correct nested wiki page instead of creating a duplicate flat one.
+- **Progress bar** for `ingest-dir` and `ingest-csv` long loops. TTY-aware
+  (in-place updates) with non-TTY fallback (periodic line output) so CI logs
+  stay readable.
+
+### Fixed
+- `mneme status` crashed with `UnboundLocalError` because a local `status` in
+  the `agent show` branch shadowed the function name throughout `main()`.
+- `wiki/HOME.md` and `wiki/<client>/HOME.md` are now skipped during HOME
+  generation so re-running is idempotent.
+
 ## [Unreleased]
 
 ### Added

{mneme_cli-0.5.0 → mneme_cli-0.5.1}/CLAUDE.md

@@ -1,4 +1,4 @@
-# Mnemosyne - Wiki Protocol
+# mneme - Wiki Protocol
 
 > **If you are an LLM agent driving mneme**, read [AGENTS.md](AGENTS.md) **first**. It is the canonical agent protocol: the agent loop, the standard task templates, the sub-agent spawning patterns, and the hard rules you must never violate. CLAUDE.md (this file) describes the wiki layer; AGENTS.md describes the agent's job.
 

{mneme_cli-0.5.0 → mneme_cli-0.5.1}/CODER.md

@@ -1,4 +1,4 @@
-# CODER.md - Developer Guide for Mnemosyne
+# CODER.md - Developer Guide for mneme
 
 This is the engineering reference for anyone building on or extending mneme. Read this before writing code.
 

{mneme_cli-0.5.0 → mneme_cli-0.5.1}/EXAMPLES.md

@@ -1,4 +1,4 @@
-# EXAMPLES.md - Mnemosyne Usage Guide
+# EXAMPLES.md - mneme Usage Guide
 
 Real-world workflows, core concepts, and entity types used in mneme.
 
@@ -256,7 +256,7 @@ cp ~/old-qms/*.pdf inbox/
 # Let tornado figure it out
 mneme tornado --client cardio-monitor
 
-#   === Mnemosyne Tornado ===
+#   === mneme Tornado ===
 #
 #   Scanning inbox/... found 50 files
 #

{mneme_cli-0.5.0 → mneme_cli-0.5.1}/FEATURES.md

@@ -1,4 +1,4 @@
-# Mnemosyne - Feature Roadmap
+# mneme - Feature Roadmap
 
 ## Current Features (v0.4.0)
 
@@ -34,6 +34,13 @@
 | `mneme tags merge` | Merge one tag into another across all pages |
 | `mneme tags suggest <page>` | Build a *tag packet* for an LLM agent (page content + taxonomy + prompt) |
 | `mneme tags apply <page> --add t1,t2 --remove t3` | Atomic tag update: rewrites frontmatter, updates schema/tags.json, re-syncs FTS5 index |
+| `mneme tags bulk-suggest --client X --filter req- --limit 50` | Bulk tag packet for many pages in one round-trip |
+| `mneme tags bulk-apply response.json` | Apply tag changes from an agent JSON response (per-page error tolerance) |
+| `mneme entity suggest --client X` | Agent-driven entity-classification packet (entities + taxonomy + example pages) |
+| `mneme entity apply --id <id> --type <type>` | Set one entity's type atomically |
+| `mneme entity bulk-apply classifications.json` | Bulk classify entities from a JSON file |
+| `mneme home --client X` / `--all-clients` | Generate a `HOME.md` navigation hub with Dataview queries and plain-markdown fallback |
+| `mneme ingest-dir --preserve-structure` | Mirror source directory hierarchy into wiki subdirectories (resync auto-detects matching subpath) |
 | `mneme diff` | Git-aware diff for a wiki page |
 | `mneme snapshot` | Versioned zip archive of a client + git tag |
 | `mneme dedupe` | Detect near-duplicate wiki pages |

{mneme_cli-0.5.0 → mneme_cli-0.5.1}/PKG-INFO

@@ -1,7 +1,7 @@
 Metadata-Version: 2.4
 Name: mneme-cli
-Version: 0.5.0
-Summary: Mnemosyne - CLI tool that turns documents into a searchable second brain. Ingest once, query forever.
+Version: 0.5.1
+Summary: mneme - CLI tool that turns documents into a searchable second brain. Ingest once, query forever.
 Author-email: Tolis Moustaklis <apostolos.moustaklis@gmail.com>
 License-Expression: MIT
 Project-URL: Homepage, https://github.com/tolism/mneme
@@ -42,7 +42,7 @@ Requires-Dist: twine>=5.0.0; extra == "release"
 Dynamic: license-file
 
 <p align="center">
-  <img src="https://raw.githubusercontent.com/tolism/mneme/main/assets/logo.png" alt="Mnemosyne" width="400">
+  <img src="https://raw.githubusercontent.com/tolism/mneme/main/assets/logo.png" alt="mneme" width="400">
 </p>
 
 <h1 align="center"></h1>
@@ -167,6 +167,13 @@ One installed CLI serves many projects — each workspace is just a directory.
 | `mneme validate writing-style <page>` | Build a *review packet* for an LLM agent to grade a page |
 | `mneme tags suggest <page>` | Build a *tag packet* for an LLM agent to choose tags |
 | `mneme tags apply <page> --add t1,t2 --remove t3` | Atomic tag update (frontmatter + schema + search index) |
+| `mneme tags bulk-suggest --client X --filter req- --limit 50` | Build one *bulk packet* covering many pages |
+| `mneme tags bulk-apply response.json` | Apply tag changes from an agent JSON response |
+| `mneme entity suggest --client X` | Build an *entity-classification packet* for an LLM agent |
+| `mneme entity apply --id <id> --type <type>` | Set one entity's type atomically |
+| `mneme entity bulk-apply classifications.json` | Bulk classify many entities |
+| `mneme home --client X` / `--all-clients` | Generate a `HOME.md` navigation hub (Dataview + fallback) |
+| `mneme ingest-dir --recursive --preserve-structure` | Mirror source directory hierarchy into the wiki |
 | `mneme agent plan --goal "..." --doc-type <t> --client <c>` | Generate a deterministic TODO plan from the active profile |
 | `mneme agent next-task` | Return the next ready task in the active plan |
 | `mneme agent task-done <id>` | Mark a task as done |

{mneme_cli-0.5.0 → mneme_cli-0.5.1}/README.md

@@ -1,5 +1,5 @@
 <p align="center">
-  <img src="https://raw.githubusercontent.com/tolism/mneme/main/assets/logo.png" alt="Mnemosyne" width="400">
+  <img src="https://raw.githubusercontent.com/tolism/mneme/main/assets/logo.png" alt="mneme" width="400">
 </p>
 
 <h1 align="center"></h1>
@@ -124,6 +124,13 @@ One installed CLI serves many projects — each workspace is just a directory.
 | `mneme validate writing-style <page>` | Build a *review packet* for an LLM agent to grade a page |
 | `mneme tags suggest <page>` | Build a *tag packet* for an LLM agent to choose tags |
 | `mneme tags apply <page> --add t1,t2 --remove t3` | Atomic tag update (frontmatter + schema + search index) |
+| `mneme tags bulk-suggest --client X --filter req- --limit 50` | Build one *bulk packet* covering many pages |
+| `mneme tags bulk-apply response.json` | Apply tag changes from an agent JSON response |
+| `mneme entity suggest --client X` | Build an *entity-classification packet* for an LLM agent |
+| `mneme entity apply --id <id> --type <type>` | Set one entity's type atomically |
+| `mneme entity bulk-apply classifications.json` | Bulk classify many entities |
+| `mneme home --client X` / `--all-clients` | Generate a `HOME.md` navigation hub (Dataview + fallback) |
+| `mneme ingest-dir --recursive --preserve-structure` | Mirror source directory hierarchy into the wiki |
 | `mneme agent plan --goal "..." --doc-type <t> --client <c>` | Generate a deterministic TODO plan from the active profile |
 | `mneme agent next-task` | Return the next ready task in the active plan |
 | `mneme agent task-done <id>` | Mark a task as done |

{mneme_cli-0.5.0 → mneme_cli-0.5.1}/mneme/__init__.py

@@ -1,8 +1,8 @@
-"""Mnemosyne - knowledge engine for documents.
+"""mneme - knowledge engine for documents.
 
 Public API:
     from mneme import core, config
     from mneme.core import ingest_source_to_both, dual_search, ...
 """
 
-__version__ = "0.5.0"
+__version__ = "0.5.1"

{mneme_cli-0.5.0 → mneme_cli-0.5.1}/mneme/config.py

@@ -1,5 +1,5 @@
 """
-Mnemosyne configuration and path resolution.
+mneme configuration and path resolution.
 
 Two distinct roots:
 
@@ -83,7 +83,7 @@ LOG_MAX_ENTRIES = 500  # archive after this many entries
 ENTITY_STOPWORDS = {
     'key facts', 'open questions', 'page title', 'page types', 'special files',
     'client directories', 'cross references', 'source summary', 'detail section',
-    'wiki page', 'wiki pages', 'wiki protocol', 'knowledge engine', 'mnemosyne', 'mneme', 'summary section',
+    'wiki page', 'wiki pages', 'wiki protocol', 'knowledge engine', 'mneme', 'summary section',
     'how to', 'last updated', 'activity log', 'health report', 'action plan',
     'executive summary', 'final verdict', 'risk scorecard', 'prior art',
     'patent strategy', 'filing strategy', 'negotiation strategy',