conduct-cli 0.4.23__tar.gz → 0.4.24__tar.gz

{conduct_cli-0.4.23 → conduct_cli-0.4.24}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: conduct-cli
-Version: 0.4.23
+Version: 0.4.24
 Summary: CLI for Conduct AI — install agents, manage projects, run tests
 Author-email: Conduct AI <hello@conductai.ai>
 License: MIT
@@ -110,13 +110,14 @@ ConductGuard is AI tool fleet management — your security team sets policies on
 ### How it works
 
 ```
-Manager installs Guard (conductai.ai/settings/modules)
-    └─ generates an invite code
+Admin configures policies and budgets in the Guard dashboard
+    └─ developers are workspace members automatically — no invite step needed
 
-Developer runs: conduct guard join <invite-code>
-    ├─ downloads team policy to ~/.conductguard/policy.json
-    ├─ writes PreToolUse hook → ~/.claude/settings.json
-    └─ registers conductguard-mcp → ~/.claude/settings.json (mcpServers)
+Developer runs: conduct guard sync
+    ├─ pulls latest policy to ~/.conductguard/policy.json
+    ├─ writes PreToolUse hook → ~/.conductguard/hook.py
+    ├─ registers hook → ~/.claude/settings.json
+    └─ registers conductguard-mcp → ~/.claude/settings.json (mcpServers) + Codex
 
 Every Claude Code tool call:
     ├─ PreToolUse hook fires (hook.py) → checks policy → block / warn / audit
@@ -126,10 +127,13 @@ Every Claude Code tool call:
 ### Developer setup
 
 ```bash
-# Get the invite code from your manager (Settings → Modules → ConductGuard)
-conduct guard join <invite-code>
+pip install conduct-cli
+
+# Authenticate (already done if you use Conduct)
+conduct login --server https://api.conductai.ai --api-key <api-key>
 
-# Enter your email when prompted — you'll be connected immediately
+# Sync Guard — installs hook + MCP, pulls policies
+conduct guard sync
 ```
 
 That's it. Policy enforcement is active from the next tool call.
@@ -138,14 +142,13 @@ That's it. Policy enforcement is active from the next tool call.
 
 | Command | Description |
 |---------|-------------|
-| `conduct guard join <code>` | Join a team, download policy, register hook + MCP |
-| `conduct guard sync` | Pull latest policy from server (run after security team updates rules) |
+| `conduct guard sync` | Pull latest policy, write hook to `~/.conductguard/hook.py`, register hook + MCP |
 | `conduct guard status` | Show today's spend, session count, and violations |
 | `conduct guard audit [--since 7d]` | Print recent guard events in a table |
 
 ### How the PreToolUse hook works
 
-When you run `conduct guard join`, the CLI writes a Python script to `~/.conductguard/hook.py` and registers it as a `PreToolUse` hook in `~/.claude/settings.json`:
+When you run `conduct guard sync`, the CLI writes a Python script to `~/.conductguard/hook.py` and registers it as a `PreToolUse` hook in `~/.claude/settings.json`:
 
 ```json
 {
@@ -173,7 +176,7 @@ Before every tool call, Claude Code runs the hook. The hook:
 
 ### How conductguard-mcp works
 
-`conduct guard join` also registers an MCP server entry in `~/.claude/settings.json`:
+`conduct guard sync` also registers an MCP server entry in `~/.claude/settings.json`:
 
 ```json
 {
@@ -244,7 +247,7 @@ Policy is stored at `~/.conductguard/policy.json` and synced from the server:
 
 ### Keeping policy up to date
 
-Policy is written to disk at `join` time. Run `conduct guard sync` after your security team updates rules in the ConductGuard dashboard. The sync command also re-registers the MCP entry in any newly detected AI tool configs.
+Run `conduct guard sync` after your security team updates rules in the ConductGuard dashboard. The sync command pulls the latest policy, rewrites the hook, and re-registers the MCP entry in any newly detected AI tool configs.
 
 ```bash
 # Add to a daily cron or run manually after policy changes

{conduct_cli-0.4.23 → conduct_cli-0.4.24}/README.md

@@ -86,13 +86,14 @@ ConductGuard is AI tool fleet management — your security team sets policies on
 ### How it works
 
 ```
-Manager installs Guard (conductai.ai/settings/modules)
-    └─ generates an invite code
+Admin configures policies and budgets in the Guard dashboard
+    └─ developers are workspace members automatically — no invite step needed
 
-Developer runs: conduct guard join <invite-code>
-    ├─ downloads team policy to ~/.conductguard/policy.json
-    ├─ writes PreToolUse hook → ~/.claude/settings.json
-    └─ registers conductguard-mcp → ~/.claude/settings.json (mcpServers)
+Developer runs: conduct guard sync
+    ├─ pulls latest policy to ~/.conductguard/policy.json
+    ├─ writes PreToolUse hook → ~/.conductguard/hook.py
+    ├─ registers hook → ~/.claude/settings.json
+    └─ registers conductguard-mcp → ~/.claude/settings.json (mcpServers) + Codex
 
 Every Claude Code tool call:
     ├─ PreToolUse hook fires (hook.py) → checks policy → block / warn / audit
@@ -102,10 +103,13 @@ Every Claude Code tool call:
 ### Developer setup
 
 ```bash
-# Get the invite code from your manager (Settings → Modules → ConductGuard)
-conduct guard join <invite-code>
+pip install conduct-cli
+
+# Authenticate (already done if you use Conduct)
+conduct login --server https://api.conductai.ai --api-key <api-key>
 
-# Enter your email when prompted — you'll be connected immediately
+# Sync Guard — installs hook + MCP, pulls policies
+conduct guard sync
 ```
 
 That's it. Policy enforcement is active from the next tool call.
@@ -114,14 +118,13 @@ That's it. Policy enforcement is active from the next tool call.
 
 | Command | Description |
 |---------|-------------|
-| `conduct guard join <code>` | Join a team, download policy, register hook + MCP |
-| `conduct guard sync` | Pull latest policy from server (run after security team updates rules) |
+| `conduct guard sync` | Pull latest policy, write hook to `~/.conductguard/hook.py`, register hook + MCP |
 | `conduct guard status` | Show today's spend, session count, and violations |
 | `conduct guard audit [--since 7d]` | Print recent guard events in a table |
 
 ### How the PreToolUse hook works
 
-When you run `conduct guard join`, the CLI writes a Python script to `~/.conductguard/hook.py` and registers it as a `PreToolUse` hook in `~/.claude/settings.json`:
+When you run `conduct guard sync`, the CLI writes a Python script to `~/.conductguard/hook.py` and registers it as a `PreToolUse` hook in `~/.claude/settings.json`:
 
 ```json
 {
@@ -149,7 +152,7 @@ Before every tool call, Claude Code runs the hook. The hook:
 
 ### How conductguard-mcp works
 
-`conduct guard join` also registers an MCP server entry in `~/.claude/settings.json`:
+`conduct guard sync` also registers an MCP server entry in `~/.claude/settings.json`:
 
 ```json
 {
@@ -220,7 +223,7 @@ Policy is stored at `~/.conductguard/policy.json` and synced from the server:
 
 ### Keeping policy up to date
 
-Policy is written to disk at `join` time. Run `conduct guard sync` after your security team updates rules in the ConductGuard dashboard. The sync command also re-registers the MCP entry in any newly detected AI tool configs.
+Run `conduct guard sync` after your security team updates rules in the ConductGuard dashboard. The sync command pulls the latest policy, rewrites the hook, and re-registers the MCP entry in any newly detected AI tool configs.
 
 ```bash
 # Add to a daily cron or run manually after policy changes

{conduct_cli-0.4.23 → conduct_cli-0.4.24}/pyproject.toml

@@ -4,7 +4,7 @@ build-backend = "setuptools.build_meta"
 
 [project]
 name = "conduct-cli"
-version = "0.4.23"
+version = "0.4.24"
 description = "CLI for Conduct AI — install agents, manage projects, run tests"
 readme = "README.md"
 license = { text = "MIT" }

{conduct_cli-0.4.23 → conduct_cli-0.4.24}/src/conduct_cli/guard.py

@@ -482,7 +482,7 @@ def _register_mcp(workspace_id: str, member_token: str, api_url: str) -> None:
     """Write conductguard MCP entry into every AI tool config found on this machine.
 
     Credentials are NOT stored in the MCP config — the server reads them from
-    ~/.conductguard/config.json at startup, which is already written by guard join.
+    ~/.conductguard/config.json at startup, which is written by guard sync.
     """
     entry = {"command": "conductguard-mcp"}
     found_any = False
@@ -840,9 +840,90 @@ def cmd_guard_sync(args):
     _register_mcp(workspace_id, cfg2.get("member_token", ""), base_url)
     print(f"  {GREEN}Hook script updated{RESET}")
 
+    # Capture savings from RTK and Agent Booster
+    _report_savings(cfg, base_url, api_key)
+
     print(f"\n{BOLD}Policy refreshed ({rule_count} rule(s)).{RESET}")
 
 
+def _report_savings(cfg: dict, base_url: str, api_key: str) -> None:
+    import subprocess
+
+    rtk_data = {}
+    booster_data = {}
+
+    # Read RTK savings — rtk gain -f json nests under "summary" key
+    try:
+        r = subprocess.run(["rtk", "gain", "-f", "json"], capture_output=True, text=True, timeout=10)
+        if r.returncode == 0:
+            raw = json.loads(r.stdout)
+            summary = raw.get("summary", raw)
+            rtk_data = {
+                "saved_tokens": summary.get("total_saved", 0),
+                "savings_pct": summary.get("avg_savings_pct", 0.0),
+                "total_commands": summary.get("total_commands", 0),
+            }
+    except Exception:
+        pass
+
+    # Read Agent Booster savings
+    try:
+        r = subprocess.run(["booster", "gain", "-f", "json"], capture_output=True, text=True, timeout=10)
+        if r.returncode == 0:
+            raw = json.loads(r.stdout)
+            booster_data = {
+                "saved_tokens": raw.get("saved_tokens", 0),
+                "savings_pct": raw.get("savings_pct", 0.0),
+                "total_reads": raw.get("total_reads", 0),
+            }
+    except Exception:
+        pass
+
+    # If neither tool returned data, skip silently
+    if not rtk_data and not booster_data:
+        return
+
+    # Load baseline to compute period_start
+    baseline_path = GUARD_DIR / "savings_baseline.json"
+    period_start = None
+    try:
+        if baseline_path.exists():
+            baseline = json.loads(baseline_path.read_text())
+            period_start = baseline.get("recorded_at")
+    except Exception:
+        pass
+
+    now_iso = datetime.now(timezone.utc).isoformat()
+
+    payload = {
+        "workspace_id": cfg.get("workspace_id", ""),
+        "member_email": cfg.get("user_email", ""),
+        "rtk": rtk_data,
+        "booster": booster_data,
+        "period_start": period_start,
+        "period_end": now_iso,
+    }
+
+    try:
+        headers = {"Content-Type": "application/json"}
+        if api_key:
+            headers["X-Api-Key"] = api_key
+        data = json.dumps(payload).encode()
+        req = urllib.request.Request(
+            f"{base_url}/guard/savings",
+            data=data,
+            headers=headers,
+            method="POST",
+        )
+        with urllib.request.urlopen(req, timeout=15) as resp:
+            resp.read()
+        # Save baseline for next diff
+        baseline_path.write_text(json.dumps({"recorded_at": now_iso, "rtk": rtk_data, "booster": booster_data}))
+        print(f"  {GREEN}Savings reported{RESET}")
+    except Exception:
+        pass  # Never fail sync because savings POST failed
+
+
 def cmd_guard_status(args):
     cfg          = _require_guard_config()
     workspace_id = cfg.get("workspace_id")

{conduct_cli-0.4.23 → conduct_cli-0.4.24}/src/conduct_cli/guardmcp.py

@@ -3,7 +3,7 @@
 conductguard-mcp — ConductGuard MCP server.
 
 Runs as a subprocess started by Claude Code / Cursor / Windsurf via the
-mcpServers config written by `conduct guard join`. Communicates over
+mcpServers config written by `conduct guard sync`. Communicates over
 stdin/stdout using JSON-RPC 2.0 (MCP stdio transport).
 
 Exposes three tools:

{conduct_cli-0.4.23 → conduct_cli-0.4.24}/src/conduct_cli.egg-info/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: conduct-cli
-Version: 0.4.23
+Version: 0.4.24
 Summary: CLI for Conduct AI — install agents, manage projects, run tests
 Author-email: Conduct AI <hello@conductai.ai>
 License: MIT
@@ -110,13 +110,14 @@ ConductGuard is AI tool fleet management — your security team sets policies on
 ### How it works
 
 ```
-Manager installs Guard (conductai.ai/settings/modules)
-    └─ generates an invite code
+Admin configures policies and budgets in the Guard dashboard
+    └─ developers are workspace members automatically — no invite step needed
 
-Developer runs: conduct guard join <invite-code>
-    ├─ downloads team policy to ~/.conductguard/policy.json
-    ├─ writes PreToolUse hook → ~/.claude/settings.json
-    └─ registers conductguard-mcp → ~/.claude/settings.json (mcpServers)
+Developer runs: conduct guard sync
+    ├─ pulls latest policy to ~/.conductguard/policy.json
+    ├─ writes PreToolUse hook → ~/.conductguard/hook.py
+    ├─ registers hook → ~/.claude/settings.json
+    └─ registers conductguard-mcp → ~/.claude/settings.json (mcpServers) + Codex
 
 Every Claude Code tool call:
     ├─ PreToolUse hook fires (hook.py) → checks policy → block / warn / audit
@@ -126,10 +127,13 @@ Every Claude Code tool call:
 ### Developer setup
 
 ```bash
-# Get the invite code from your manager (Settings → Modules → ConductGuard)
-conduct guard join <invite-code>
+pip install conduct-cli
+
+# Authenticate (already done if you use Conduct)
+conduct login --server https://api.conductai.ai --api-key <api-key>
 
-# Enter your email when prompted — you'll be connected immediately
+# Sync Guard — installs hook + MCP, pulls policies
+conduct guard sync
 ```
 
 That's it. Policy enforcement is active from the next tool call.
@@ -138,14 +142,13 @@ That's it. Policy enforcement is active from the next tool call.
 
 | Command | Description |
 |---------|-------------|
-| `conduct guard join <code>` | Join a team, download policy, register hook + MCP |
-| `conduct guard sync` | Pull latest policy from server (run after security team updates rules) |
+| `conduct guard sync` | Pull latest policy, write hook to `~/.conductguard/hook.py`, register hook + MCP |
 | `conduct guard status` | Show today's spend, session count, and violations |
 | `conduct guard audit [--since 7d]` | Print recent guard events in a table |
 
 ### How the PreToolUse hook works
 
-When you run `conduct guard join`, the CLI writes a Python script to `~/.conductguard/hook.py` and registers it as a `PreToolUse` hook in `~/.claude/settings.json`:
+When you run `conduct guard sync`, the CLI writes a Python script to `~/.conductguard/hook.py` and registers it as a `PreToolUse` hook in `~/.claude/settings.json`:
 
 ```json
 {
@@ -173,7 +176,7 @@ Before every tool call, Claude Code runs the hook. The hook:
 
 ### How conductguard-mcp works
 
-`conduct guard join` also registers an MCP server entry in `~/.claude/settings.json`:
+`conduct guard sync` also registers an MCP server entry in `~/.claude/settings.json`:
 
 ```json
 {
@@ -244,7 +247,7 @@ Policy is stored at `~/.conductguard/policy.json` and synced from the server:
 
 ### Keeping policy up to date
 
-Policy is written to disk at `join` time. Run `conduct guard sync` after your security team updates rules in the ConductGuard dashboard. The sync command also re-registers the MCP entry in any newly detected AI tool configs.
+Run `conduct guard sync` after your security team updates rules in the ConductGuard dashboard. The sync command pulls the latest policy, rewrites the hook, and re-registers the MCP entry in any newly detected AI tool configs.
 
 ```bash
 # Add to a daily cron or run manually after policy changes