sphica 0.1.0 → 0.3.0

package/.claude-plugin/plugin.json

@@ -1,7 +1,7 @@
 {
   "$schema": "https://anthropic.com/claude-code/plugin.schema.json",
   "name": "sphica",
-  "version": "0.1.0",
+  "version": "0.3.0",
   "description": "Records conversations automatically and recalls past decisions, rejected options, constraints, and what was said.",
   "author": {
     "name": "iroha924",

package/.codex-plugin/plugin.json

@@ -1,6 +1,6 @@
 {
   "name": "sphica",
-  "version": "0.1.0",
+  "version": "0.3.0",
   "description": "Records conversations automatically and recalls past decisions, rejected options, constraints, and what was said.",
   "author": {
     "name": "iroha924",

package/README.md

@@ -21,7 +21,6 @@ The database is a single SQLite file on your machine.
 - **Automatic session recording.** Sphica keeps your prompts, the agent's final reply for each turn, and the paths of files changed by the agent's edit tools.
 - **Decision records on request.** `/sphica:trace` saves the decisions, rejected options, constraints, and dead ends of a session, plus where the work stands.
 - **Multi-perspective review.** `/sphica:review` runs a separate reviewer for each focus: correctness, security, and written conventions by default, plus redundancy and past decisions with `full`. When Codex is installed, it offers to repeat the review with Codex.
-- **Terminal dashboard.** `sphica dashboard` browses sessions, work in progress, and search results. It is read-only.
 - **GitHub and docs import.** `sphica harvest` imports pull requests, issues, and the repository's Markdown files.
 
 The agent is told to treat records as history, not instructions, and to trust the code when a record and the current code disagree.
@@ -61,13 +60,14 @@ codex plugin add sphica@sphica
 
 In Codex, open `/hooks` and mark Sphica's hooks as trusted. Nothing is recorded until you do. If a plugin update changes the hooks, trust them again.
 
-**3. Create the database**
+**3. Set up in your repository**
 
 ```bash
+cd ~/Projects/your-repo
 sphica init
 ```
 
-This creates `~/.sphica/sphica.db`. Running it again leaves an existing database untouched.
+This creates `~/.sphica/sphica.db` and registers the repository. Running it again leaves both untouched. If the repository has no `origin` remote, give it a name: `sphica init --name <name>`. Add `--sync` to import its GitHub history and docs right away.
 
 **4. Check the setup**
 
@@ -79,14 +79,7 @@ sphica doctor
 
 ## Quick start
 
-Sphica writes sessions to the database only for repositories you register. A registered repository is called a project.
-
-```bash
-cd ~/Projects/your-repo
-sphica project add
-```
-
-If the repository has no `origin` remote, give it a name: `sphica project add --name <name>`.
+Sphica writes sessions to the database only for repositories you register. A registered repository is called a project; run `sphica init` in each repository you want recorded.
 
 Then work as usual in Claude Code or Codex. To bring back earlier decisions, ask the agent:
 
@@ -107,12 +100,6 @@ sphica harvest --cwd .      # only the current repository
 Without `--cwd`, Sphica looks for projects directly under `~/Projects` and for projects registered with `--name`. Use `--cwd` for a repository somewhere else.
 Docs are read from the default branch of `origin`, or from the local `HEAD` when there is no `origin`.
 
-To browse everything in the terminal:
-
-```bash
-sphica dashboard   # Tab switches screens, / searches, p changes project, q quits
-```
-
 ## What gets recorded and where it goes
 
 - **Where.** The database is `~/.sphica/sphica.db`. Records wait in a local queue, `~/.sphica/spool`, until they are written to it. Each machine has its own database, and records are not shared between machines.
@@ -190,14 +177,12 @@ Run `sphica doctor` first. It shows which part is out of date or not working. Co
 
 | Command | What it does |
 |---|---|
-| `sphica init` | Create the database |
-| `sphica doctor` | Check versions, the database, and recording |
-| `sphica project add` | Register the current repository as a project |
+| `sphica init` | Create the database and register the current repository |
+| `sphica doctor` | Check versions, the database, recording, and each project's last import |
 | `sphica harvest` | Import GitHub pull requests, issues, and Markdown docs |
-| `sphica search <words>` | Search from the terminal |
-| `sphica dashboard` | Browse sessions, work, and search results |
+| `sphica advice` | See how often the edit hook showed constraints |
 
-Run `sphica --help` for the full list and `sphica <command> --help` for each command's options.
+Run `sphica --help` for these, `sphica -H` for every command (managing projects and people, and the ones agents and maintenance use), and `sphica <command> --help` for each command's options.
 
 ## Security