agenr 0.4.1 → 0.5.2

package/CHANGELOG.md

@@ -2,19 +2,79 @@
 
 ## Unreleased
 
+### Added
+- `entries.project` column (with index) to tag knowledge by source project/repo (NULL for legacy entries)
+- Project auto-detection from transcript CWD in watch mode (tags entries at write time)
+- `--project` and `--exclude-project` filters/tags across commands:
+  - `agenr recall --project/--exclude-project [--strict]`
+  - `agenr context --project/--exclude-project [--strict]`
+  - `agenr store --project`
+  - `agenr ingest --project`
+  - `agenr consolidate --project/--exclude-project` (never merges across projects)
+  - `agenr db stats --project/--exclude-project`
+  - `agenr db export --project/--exclude-project`
+- MCP tool support for project:
+  - `agenr_recall` accepts optional `project` filter (comma-separated for multiple)
+  - `agenr_store` accepts optional `project` tag
+- `agenr eval recall` command for scoring regression checks (baseline save and compare)
+
+### Fixed
+- Recall scoring and session-start recall:
+  - Freshness boost for importance >= 6 (clamped to avoid amplifying noisy entries)
+  - Smooth exponential todo staleness decay (half-life 7 days; floors at 0.10 or 0.40 for importance >= 8)
+  - Session-start permanent window widened to 30 days (temporary remains shorter)
+  - Dynamic budget allocation based on available categories
+  - Recency tiebreaking within a 0.05 score dead-band applied to the recent category only
+- Watch ingestion now advances `byteOffset` by bytes actually read in each cycle, preventing duplicate processing when files grow during read.
+- Watch state saves are now atomic (temp file + rename), preventing partial-write corruption on process crashes.
+
+## 0.5.0 (2026-02-17)
+
 ### Added
 - `_meta` table with schema version stamp for future migrations
 - `agenr db version` command to print schema version metadata
+- `agenr daemon start|stop|restart` commands
+- `agenr daemon install --dir/--platform/--node-path` options for explicit daemon configuration
+- `entries.platform` column (with index) to tag knowledge by platform (`openclaw|claude-code|codex`, NULL for legacy entries)
+- `--platform` filters/tags across commands:
+  - `agenr recall --platform`
+  - `agenr context --platform`
+  - `agenr store --platform`
+  - `agenr ingest --platform`
+  - `agenr consolidate --platform`
+  - `agenr db export --platform`
+- MCP tool support for platform:
+  - `agenr_recall` accepts optional `platform` filter
+  - `agenr_store` accepts optional `platform` tag
 
 ### Changed
 - `agenr db stats` output now includes schema version
+- `agenr db stats` now includes per-platform breakdown
+- `agenr daemon install` now uses smart platform defaults and writes `watch --dir <path> --platform <name>` instead of `watch --auto`
+- `agenr daemon install` now prefers stable node symlinks (Homebrew) when `process.execPath` is version-specific; use `--node-path` to override
+- `agenr watch --auto` is deprecated; `agenr watch --platform <name>` is now the standard invocation and auto-resolves the default platform directory when `--dir` is omitted
+
+## 0.4.1 (2026-02-17)
 
-## 0.4.0
+### Fixed
+- npx symlink handling: isDirectRun check now uses realpathSync to resolve npx symlinks correctly
+
+## 0.4.0 (2026-02-15)
 
 ### Added
 - `agenr context` command - generate context files for AI tool integration
 - `agenr watch --context` - auto-refresh context file after each extraction cycle
 - `agenr daemon` - launchd daemon management for background watching
+- `agenr consolidate` - knowledge base cleanup with rule-based and LLM-assisted merging
+- Online dedup at write time (mem0-style dedup with 3 cosine bands)
+- Post-extraction LLM dedup pass
+- Concurrent chunk extraction
+- Smart filtering before chunking
+- Rate limit protection for chunk extraction
+- Graceful shutdown for long-running commands (SIGINT/SIGTERM)
+- Ingest auto-retry for failed files
+- Source adapter refactor with timestamp preservation
+- Watch WAL checkpointing
 
 ### Changed
 - Embedding dimensions upgraded from 512 to 1024 (text-embedding-3-small)
@@ -24,26 +84,21 @@
 - Session-start recall no longer dominated by stale todos (todo staleness penalty)
 - Consolidate releases DB lock after WAL checkpoint, not before
 
-## 0.3.1
-
-### Added
-- `agenr consolidate` - knowledge base cleanup with rule-based and LLM-assisted merging
-
-## 0.3.0
+## 0.3.0 (2026-02-15)
 
 ### Added
 - `agenr watch` - live file watcher with auto-extraction
 - `agenr ingest` - bulk ingestion of markdown, plaintext, and JSONL
 - `agenr mcp` - MCP server for cross-tool AI memory (recall, store, extract)
 
-## 0.2.0
+## 0.2.0 (2026-02-14)
 
 ### Added
 - `agenr store` - smart dedup with cosine similarity bands
 - `agenr recall` - recall with scoring and budget-constrained retrieval
 - `agenr db` subcommands (stats, export, reset, path)
 
-## 0.1.0
+## 0.1.0 (2026-02-14)
 
 ### Added
 - `agenr extract` - structured knowledge extraction from conversation transcripts

package/README.md

@@ -1,8 +1,8 @@
-# agenr
+# AGENR
 
-**/eɪ.dʒɛn.ɚ/** (AY-GEN-ER) - local-first memory for AI agents.
+**AY-JEN-ER** - Human memory for AI agents.
 
-Your AI forgets everything between sessions. agenr fixes that.
+Your AI forgets everything between sessions. AGENR fixes that.
 
 It extracts structured knowledge from your OpenClaw conversation transcripts - facts, decisions, preferences, todos, relationships, events, lessons - and stores them in a local database with semantic search. Entries strengthen when reinforced, decay when stale, and resolve contradictions. It's not a search engine. It's memory that gets healthier with use.
 
@@ -10,9 +10,9 @@ One local database. Your memory stays on your machine.
 
 ## What you need
 
-agenr uses embeddings to make your memory searchable. The best setup we've found: an **OpenAI API key** with `text-embedding-3-small`. Embeddings cost fractions of a penny per operation - a full ingestion of 100+ session transcripts runs about $0.10 total.
+AGENR uses embeddings to make your memory searchable. The best setup we've found: an **OpenAI API key** with `text-embedding-3-small`. Embeddings cost fractions of a penny per operation - a full ingestion of 100+ session transcripts runs about $0.10 total.
 
-agenr also supports **OpenAI Pro subscriptions** and **Anthropic Claude subscriptions** (no API key needed) for the LLM extraction step. But for the best balance of speed, accuracy, and cost, we recommend `gpt-4o-mini` with an API key. `agenr setup` walks you through all of this.
+AGENR also supports **OpenAI Pro subscriptions** and **Anthropic Claude subscriptions** (no API key needed) for the LLM extraction step. But for the best balance of speed, accuracy, and cost, we recommend `gpt-4o-mini` with an API key. `agenr setup` walks you through all of this.
 
 ```bash
 export OPENAI_API_KEY=sk-...  # for embeddings + extraction
@@ -85,7 +85,7 @@ Done. Your agent now has persistent memory that survives compaction, session res
 
 ## What happens when you ingest
 
-agenr reads your OpenClaw session transcripts, filters out noise (tool calls, file dumps, boilerplate - about 80% of a typical session), and extracts structured knowledge entries:
+AGENR reads your OpenClaw session transcripts, filters out noise (tool calls, file dumps, boilerplate - about 80% of a typical session), and extracts structured knowledge entries:
 
 ```
 agenr ingest ~/.openclaw/agents/main/sessions/ --glob '**/*.jsonl'
@@ -113,10 +113,10 @@ agenr recall "package manager"
 The watcher keeps your memory current as you work. It tails your session files, extracts new knowledge every few minutes, and stores it. If you ingested history first, watch resumes right where ingest left off - no re-processing.
 
 ```bash
-# Watch your OpenClaw sessions directory
-agenr watch --dir ~/.openclaw/agents/main/sessions/
+# Watch your OpenClaw sessions directory (auto-resolves the default path)
+agenr watch --platform openclaw
 
-# Auto-detect your session directory
+# Deprecated: --auto still works (defaults to OpenClaw) but will be removed in a future version
 agenr watch --auto
 
 # Install as a background daemon (macOS launchd)
@@ -128,7 +128,7 @@ agenr daemon logs
 You can also auto-refresh a context file that AI tools read on startup:
 
 ```bash
-agenr watch --auto --context ~/.agenr/CONTEXT.md
+agenr watch --platform openclaw --context ~/.agenr/CONTEXT.md
 ```
 
 ## How it works
@@ -150,7 +150,7 @@ Transcript -> Filter -> Extract -> Store -> Recall
 
 ## MCP integration
 
-agenr exposes three MCP tools: `agenr_recall`, `agenr_store`, `agenr_extract`. Any tool that speaks MCP can use your memory.
+AGENR exposes three MCP tools: `agenr_recall`, `agenr_store`, `agenr_extract`. Any tool that speaks MCP can use your memory.
 
 **OpenClaw** (via [mcporter](https://mcporter.dev)):
 ```bash
@@ -164,6 +164,8 @@ Verify with `mcporter list agenr`. See [docs/OPENCLAW.md](./docs/OPENCLAW.md) fo
 | Command | What it does |
 | --- | --- |
 | `agenr setup` | Interactive configuration |
+| `agenr config` | Show and update agenr configuration |
+| `agenr auth` | Authentication status and diagnostics |
 | `agenr ingest <paths...>` | Bulk-ingest files and directories |
 | `agenr extract <files...>` | Extract knowledge from text |
 | `agenr store [files...]` | Store entries with semantic dedup |
@@ -173,8 +175,7 @@ Verify with `mcporter list agenr`. See [docs/OPENCLAW.md](./docs/OPENCLAW.md) fo
 | `agenr consolidate` | Clean up and merge near-duplicates |
 | `agenr context` | Generate context file for AI tool integration |
 | `agenr mcp` | Start MCP server (stdio) |
-| `agenr db stats` | Database statistics |
-| `agenr db version` | Database schema version |
+| `agenr db <cmd>` | Database management (stats, version, export, reset, path, check, rebuild-index). Run `agenr db --help` for all subcommands. |
 
 Full reference: [docs/CLI.md](./docs/CLI.md) | [docs/CONFIGURATION.md](./docs/CONFIGURATION.md)
 
@@ -189,7 +190,7 @@ Deep dive: [docs/ARCHITECTURE.md](./docs/ARCHITECTURE.md)
 
 ## Status
 
-Alpha. The core pipeline is stable and tested (369 tests). We use it daily managing thousands of knowledge entries across OpenClaw sessions.
+Alpha. The core pipeline is stable and tested (445 tests). We use it daily managing thousands of knowledge entries across OpenClaw sessions.
 
 What works: extraction, storage, recall, MCP integration, online dedup, consolidation, smart filtering, live watching, daemon mode.
 
@@ -201,7 +202,7 @@ The big labs are building bigger brains. We're building better memory. Those are
 
 Current AI's bottleneck isn't intelligence - it's continuity. A slightly less brilliant model with accumulated context might be more useful than a brilliant amnesiac. What makes a senior engineer senior isn't raw IQ - it's patterns seen, mistakes remembered, approaches that worked. That's memory.
 
-agenr is local-first because your memory is yours. It's structured (not just vectors) because "what did we decide about X?" needs a real answer, not a similarity score. It's open source because memory infrastructure should be shared.
+AGENR keeps your memory on your machine because it is yours. It's structured (not just vectors) because "what did we decide about X?" needs a real answer, not a similarity score. It's open source because memory infrastructure should be shared.
 
 ## Troubleshooting
 

package/dist/cli-main.d.ts

@@ -4,9 +4,11 @@ import { Model, Api, Context, SimpleStreamOptions, AssistantMessageEvent, Assist
 declare const KNOWLEDGE_TYPES: readonly ["fact", "decision", "preference", "todo", "relationship", "event", "lesson"];
 declare const EXPIRY_LEVELS: readonly ["core", "permanent", "temporary"];
 declare const SCOPE_LEVELS: readonly ["private", "personal", "public"];
+declare const KNOWLEDGE_PLATFORMS: readonly ["openclaw", "claude-code", "codex"];
 type KnowledgeType = (typeof KNOWLEDGE_TYPES)[number];
 type Expiry = (typeof EXPIRY_LEVELS)[number];
 type Scope = (typeof SCOPE_LEVELS)[number];
+type KnowledgePlatform = (typeof KNOWLEDGE_PLATFORMS)[number];
 type AgenrProvider = "anthropic" | "openai" | "openai-codex";
 type AgenrAuthMethod = "anthropic-oauth" | "anthropic-token" | "anthropic-api-key" | "openai-subscription" | "openai-api-key";
 interface AgenrStoredCredentials {
@@ -34,6 +36,8 @@ interface KnowledgeEntry {
     content: string;
     subject: string;
     canonical_key?: string;
+    platform?: KnowledgePlatform;
+    project?: string;
     importance: number;
     expiry: Expiry;
     scope?: Scope;