humanbound-cli 0.7.0__tar.gz → 0.9.0__tar.gz

{humanbound_cli-0.7.0 → humanbound_cli-0.9.0}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: humanbound-cli
-Version: 0.7.0
+Version: 0.9.0
 Summary: Humanbound CLI - command line interface for AI agent security testing.
 Author-email: Kostas Siabanis <hello@humanbound.ai>, Demetris Gerogiannis <hello@humanbound.ai>
 License: Apache-2.0
@@ -60,6 +60,7 @@ Humanbound runs automated adversarial attacks against your bot's live endpoint,
 | **Posture Scoring** | Quantified 0-100 security score with breakdown by findings, coverage, and resilience. Track over time. |
 | **Shadow AI Discovery** | Scan cloud tenants for AI services, assess risk with 15 SAI threat classes, and govern your AI inventory. |
 | **Guardrails Export** | Generate protection rules from test findings. Export to OpenAI or Humanbound format. |
+| **Firewall Training** | Train agent-specific Tier 2 classifiers from adversarial + QA test data. Pluggable model architecture via AgentClassifier scripts. |
 | **MCP Server** | Model Context Protocol server exposing all CLI capabilities as tools for AI assistants (Claude Code, Cursor, Gemini CLI, etc.). |
 
 ### Why Humanbound?
@@ -557,6 +558,30 @@ hb findings [--status open] [--severity high] [--json]
 hb guardrails [--vendor humanbound|openai] [--format json|yaml] [-o FILE]
 ```
 
+### Firewall
+
+Train agent-specific classifiers for [hb-firewall](https://github.com/humanbound/firewall).
+
+```bash
+# Train from adversarial + QA test data
+hb firewall train --model detectors/setfit_classifier.py
+
+# Show model info
+hb firewall show firewall.hbfw
+```
+
+| Flag | Default | Description |
+|------|---------|-------------|
+| `--model PATH` | — | Path to AgentClassifier script (required) |
+| `--last N` | 10 | Last N finished experiments |
+| `--from DATE` | — | Start date filter |
+| `--until DATE` | — | End date filter |
+| `--min-samples` | 30 | Minimum conversations required |
+| `--output` | firewall_\<project\>.hbfw | Output file path |
+| `--benign-dataset` | — | HuggingFace dataset for benign benchmarking |
+
+See [hb-firewall docs](https://github.com/humanbound/firewall) for the AgentClassifier interface and full integration guide.
+
 ### Shell Completion
 
 ```bash
@@ -723,6 +748,22 @@ hb connect -e ./bot-config.json
 hb test -e ./bot-config.json
 ```
 
+### Whitebox testing with telemetry
+
+Add a `telemetry` block to your agent config to enable whitebox testing. Humanbound fetches tool calls, memory operations, and resource usage from your observability platform (LangFuse, LangSmith, OpenAI Assistants, W&B, Helicone, AgentOps, or custom).
+
+```json
+{
+  "telemetry": {
+    "format": "langfuse",
+    "endpoint": "https://cloud.langfuse.com/api/public/sessions/$session_id",
+    "headers": { "Authorization": "Basic <base64(pk:sk)>" }
+  }
+}
+```
+
+See the full [Telemetry Integration Guide](https://docs.humanbound.ai/integrations/telemetry/) for vendor-specific setup and the custom extraction map reference.
+
 ### Shadow AI discovery & governance
 
 ```bash
@@ -759,6 +800,29 @@ claude mcp add humanbound -- hb mcp
 hb guardrails --vendor openai --format json -o guardrails.json
 ```
 
+### Train and deploy firewall
+
+```bash
+# Run adversarial tests
+hb test
+
+# Train Tier 1 classifier from results
+hb firewall train -o model.hbfw
+
+# Verify quality
+hb firewall show model.hbfw
+# F1=0.95, Precision=0.97, Tier 1 coverage=92%
+
+# Test before deploying
+# Deploy in your app
+python -c "
+from hb_firewall import Firewall
+fw = Firewall.from_config('agent.yaml', model_path='model.hbfw')
+result = fw.evaluate('What is my balance?')
+print(result.verdict, result.tier)  # Verdict.PASS 1
+"
+```
+
 ---
 
 ### On-Premises

{humanbound_cli-0.7.0 → humanbound_cli-0.9.0}/README.md

@@ -27,6 +27,7 @@ Humanbound runs automated adversarial attacks against your bot's live endpoint,
 | **Posture Scoring** | Quantified 0-100 security score with breakdown by findings, coverage, and resilience. Track over time. |
 | **Shadow AI Discovery** | Scan cloud tenants for AI services, assess risk with 15 SAI threat classes, and govern your AI inventory. |
 | **Guardrails Export** | Generate protection rules from test findings. Export to OpenAI or Humanbound format. |
+| **Firewall Training** | Train agent-specific Tier 2 classifiers from adversarial + QA test data. Pluggable model architecture via AgentClassifier scripts. |
 | **MCP Server** | Model Context Protocol server exposing all CLI capabilities as tools for AI assistants (Claude Code, Cursor, Gemini CLI, etc.). |
 
 ### Why Humanbound?
@@ -524,6 +525,30 @@ hb findings [--status open] [--severity high] [--json]
 hb guardrails [--vendor humanbound|openai] [--format json|yaml] [-o FILE]
 ```
 
+### Firewall
+
+Train agent-specific classifiers for [hb-firewall](https://github.com/humanbound/firewall).
+
+```bash
+# Train from adversarial + QA test data
+hb firewall train --model detectors/setfit_classifier.py
+
+# Show model info
+hb firewall show firewall.hbfw
+```
+
+| Flag | Default | Description |
+|------|---------|-------------|
+| `--model PATH` | — | Path to AgentClassifier script (required) |
+| `--last N` | 10 | Last N finished experiments |
+| `--from DATE` | — | Start date filter |
+| `--until DATE` | — | End date filter |
+| `--min-samples` | 30 | Minimum conversations required |
+| `--output` | firewall_\<project\>.hbfw | Output file path |
+| `--benign-dataset` | — | HuggingFace dataset for benign benchmarking |
+
+See [hb-firewall docs](https://github.com/humanbound/firewall) for the AgentClassifier interface and full integration guide.
+
 ### Shell Completion
 
 ```bash
@@ -690,6 +715,22 @@ hb connect -e ./bot-config.json
 hb test -e ./bot-config.json
 ```
 
+### Whitebox testing with telemetry
+
+Add a `telemetry` block to your agent config to enable whitebox testing. Humanbound fetches tool calls, memory operations, and resource usage from your observability platform (LangFuse, LangSmith, OpenAI Assistants, W&B, Helicone, AgentOps, or custom).
+
+```json
+{
+  "telemetry": {
+    "format": "langfuse",
+    "endpoint": "https://cloud.langfuse.com/api/public/sessions/$session_id",
+    "headers": { "Authorization": "Basic <base64(pk:sk)>" }
+  }
+}
+```
+
+See the full [Telemetry Integration Guide](https://docs.humanbound.ai/integrations/telemetry/) for vendor-specific setup and the custom extraction map reference.
+
 ### Shadow AI discovery & governance
 
 ```bash
@@ -726,6 +767,29 @@ claude mcp add humanbound -- hb mcp
 hb guardrails --vendor openai --format json -o guardrails.json
 ```
 
+### Train and deploy firewall
+
+```bash
+# Run adversarial tests
+hb test
+
+# Train Tier 1 classifier from results
+hb firewall train -o model.hbfw
+
+# Verify quality
+hb firewall show model.hbfw
+# F1=0.95, Precision=0.97, Tier 1 coverage=92%
+
+# Test before deploying
+# Deploy in your app
+python -c "
+from hb_firewall import Firewall
+fw = Firewall.from_config('agent.yaml', model_path='model.hbfw')
+result = fw.evaluate('What is my balance?')
+print(result.verdict, result.tier)  # Verdict.PASS 1
+"
+```
+
 ---
 
 ### On-Premises

{humanbound_cli-0.7.0 → humanbound_cli-0.9.0}/humanbound_cli/commands/__init__.py

@@ -5,7 +5,7 @@ from . import (
     guardrails, docs, providers, findings, api_keys, members,
     coverage, campaigns, upload_logs, sentinel, discover,
     connectors, inventory, completion, connect, report, monitor,
-    webhooks, assessments,
+    webhooks, assessments, firewall,
 )
 
 # MCP command is optional — only available when mcp SDK is installed
@@ -42,5 +42,6 @@ __all__ = [
     "monitor",
     "webhooks",
     "assessments",
+    "firewall",
     "mcp",
 ]

humanbound_cli-0.9.0/humanbound_cli/commands/firewall.py

@@ -0,0 +1,270 @@
+"""Firewall commands — train and manage Tier 2 classifiers for hb-firewall."""
+
+import sys
+import time
+from pathlib import Path
+
+import click
+from rich.console import Console
+from rich.progress import Progress
+
+from ..client import HumanboundClient
+from ..exceptions import NotAuthenticatedError, APIError
+
+console = Console()
+
+
+@click.group("firewall")
+def firewall_group():
+    """Train and manage Tier 2 classifiers for hb-firewall.
+
+    \b
+    Workflow:
+      1. Run adversarial tests:  hb test
+      2. Train classifiers:      hb firewall train --model detectors/my_model.py
+      3. Load in your app:       Firewall.from_config("agent.yaml", model_path="fw.hbfw")
+    """
+
+
+@firewall_group.command("train")
+@click.option("--model", "model_path", type=str, required=False, default=None,
+              help="Path to AgentClassifier script (e.g. detectors/one_class_svm.py)")
+@click.option("--last", "last_n", type=int, default=10,
+              help="Last N finished experiments (default: 10)")
+@click.option("--from", "from_date", type=str, default=None)
+@click.option("--until", "until_date", type=str, default=None)
+@click.option("--min-samples", type=int, default=30)
+@click.option("--output", "-o", type=click.Path(), default=None)
+def train_command(model_path, last_n, from_date, until_date, min_samples,
+                  output):
+    """Train Tier 2 classifiers from adversarial + QA test logs."""
+    if not model_path:
+        # Default to SetFit classifier shipped with hb-firewall
+        import importlib.resources
+        try:
+            # Try to find setfit_classifier.py relative to hb_firewall package
+            import hb_firewall
+            pkg_dir = Path(hb_firewall.__file__).parent.parent.parent
+            default = pkg_dir / "detectors" / "setfit_classifier.py"
+            if default.exists():
+                model_path = str(default)
+            else:
+                console.print("[red]Default SetFit classifier not found.[/red]")
+                console.print("  Provide a path to an AgentClassifier script:")
+                console.print("  hb firewall train --model detectors/setfit_classifier.py")
+                sys.exit(1)
+        except Exception:
+            console.print("[red]Provide --model flag.[/red]")
+            console.print("  hb firewall train --model detectors/setfit_classifier.py")
+            sys.exit(1)
+
+    try:
+        client = HumanboundClient()
+        project_id = client.project_id
+        if not project_id:
+            console.print("[red]No active project.[/red] Run: hb projects use <id>")
+            sys.exit(1)
+
+        # Load detector
+        try:
+            from hb_firewall.hbfw import HBFW, load_model_class, save_hbfw
+        except ImportError:
+            console.print("[red]Install: pip install hb-firewall[/red]")
+            sys.exit(1)
+
+        try:
+            detector_cls = load_model_class(model_path)
+        except ValueError as e:
+            console.print(f"[red]{e}[/red]")
+            sys.exit(1)
+
+        console.print(f"\n[bold]Training firewall classifiers[/bold]")
+        console.print(f"  Project: {project_id[:8]}...")
+        console.print(f"  Model: {model_path}")
+
+        # Step 1: Fetch experiments
+        console.print(f"\n[bold]Step 1:[/bold] Fetching experiments...")
+        adv_exps = _fetch_experiments(client, "adversarial", last_n=last_n,
+                                       from_date=from_date, until_date=until_date)
+        qa_exps = _fetch_experiments(client, "adversarial", exclude=True,
+                                      last_n=last_n, from_date=from_date, until_date=until_date)
+        if not adv_exps:
+            console.print("[red]No finished adversarial experiments.[/red] Run: hb test")
+            sys.exit(1)
+        console.print(f"  Found {len(adv_exps)} adversarial + {len(qa_exps)} QA experiments")
+
+        # Step 2: Fetch logs
+        console.print(f"\n[bold]Step 2:[/bold] Pulling conversation logs...")
+        logs = _fetch_logs(client, adv_exps + qa_exps)
+        if len(logs) < min_samples:
+            console.print(f"[red]Only {len(logs)} conversations (min: {min_samples}).[/red]")
+            sys.exit(1)
+
+        n_pass = sum(1 for l in logs if l.get("result") == "pass")
+        n_fail = len(logs) - n_pass
+        adv_fail = sum(1 for l in logs
+                       if "adversarial" in (l.get("test_category") or "")
+                       and l.get("result") == "fail")
+        console.print(f"  Collected {len(logs)} conversations ({n_pass} pass, {n_fail} fail)")
+        console.print(f"  Training on {adv_fail} failed adversarial conversations")
+
+        permitted, restricted = _fetch_intents(client, project_id)
+        if permitted or restricted:
+            console.print(f"  Intents: {len(permitted or [])} permitted, {len(restricted or [])} restricted")
+
+        # Step 3: Prepare
+        console.print(f"\n[bold]Step 3:[/bold] Preparing training data...")
+        hbfw = HBFW(attack_detector=detector_cls("attack"),
+                     benign_detector=detector_cls("benign"))
+        data = hbfw.prepare(logs, restricted_intents=restricted, permitted_intents=permitted)
+
+        stats = data.get("stats", {})
+        console.print(f"  Attack samples: {stats.get('attack_samples', 0)} (curated: {stats.get('curated_attack', 0)})")
+        console.print(f"  Benign samples: {stats.get('benign_samples', 0)} (curated: {stats.get('curated_benign', 0)})")
+        if not data.get("has_qa"):
+            console.print(f"  [yellow]No QA tests. Using permitted intents as benign data.[/yellow]")
+
+        # Step 4: Train
+        console.print(f"\n[bold]Step 4:[/bold] Training...")
+        performance = hbfw.train(data, permitted_intents=permitted,
+                                  restricted_intents=restricted)
+
+        val = performance.get("validation")
+        if val:
+            af = val.get("adversarial_fail", {})
+            ap = val.get("adversarial_pass", {})
+            bn = val.get("benign", {})
+            console.print(f"\n  [bold]Validation (conversation replay)[/bold]")
+            if af.get("total"):
+                r = af.get("rate", 0)
+                s = "green" if r >= 0.8 else "yellow" if r >= 0.5 else "red"
+                console.print(f"    Failed adversarial caught: [{s}]{af['caught']}/{af['total']} ({r:.1%})[/{s}]")
+            if ap.get("total"):
+                r = ap.get("rate", 0)
+                s = "green" if r >= 0.5 else "dim"
+                console.print(f"    Passed adversarial caught: [{s}]{ap['caught']}/{ap['total']} ({r:.1%})[/{s}]")
+            if bn.get("total"):
+                r = bn.get("rate", 0)
+                s = "green" if r >= 0.8 else "yellow" if r >= 0.5 else "red"
+                console.print(f"    Benign allowed: [{s}]{bn['correct']}/{bn['total']} ({r:.1%})[/{s}]")
+                if bn.get("blocked"):
+                    console.print(f"    [red]Benign blocked: {bn['blocked']}[/red]")
+
+        console.print(f"  Training complete.")
+
+        # Save
+        if output is None:
+            output = f"firewall_{project_id[:8]}.hbfw"
+        model_data = hbfw.export()
+        model_data["config"]["project_id"] = project_id
+        model_data["config"]["created_at"] = time.strftime("%Y-%m-%dT%H:%M:%SZ")
+        model_data["config"]["n_conversations"] = len(logs)
+        model_data["config"]["detector"] = Path(model_path).stem
+        save_hbfw(model_data, output)
+
+        file_size = Path(output).stat().st_size / 1024
+        console.print(f"\n[green]Model saved:[/green] {output} ({file_size:.0f} KB)")
+
+    except NotAuthenticatedError:
+        console.print("[red]Not authenticated.[/red] Run: hb login")
+        sys.exit(1)
+    except APIError as e:
+        console.print(f"[red]API error:[/red] {e}")
+        sys.exit(1)
+
+
+@firewall_group.command("show")
+@click.argument("model_path", type=click.Path(exists=True))
+def show_command(model_path):
+    """Show model info from a trained .hbfw file."""
+    try:
+        from hb_firewall.hbfw import load_hbfw
+    except ImportError:
+        console.print("[red]hb-firewall not installed.[/red]")
+        sys.exit(1)
+
+    config, _ = load_hbfw(model_path)
+    console.print(f"\n[bold]Firewall Model: {model_path}[/bold]")
+    console.print(f"  Created: {config.get('created_at', '?')}")
+    console.print(f"  Project: {config.get('project_id', '?')}")
+    console.print(f"  Detector: {config.get('detector', '?')}")
+    perf = config.get("performance", {})
+    stats = perf.get("stats", {})
+    if stats:
+        console.print(f"  Attack samples: {stats.get('attack_samples', '?')}")
+        console.print(f"  Benign samples: {stats.get('benign_samples', '?')}")
+    val = perf.get("validation")
+    if val:
+        af = val.get("adversarial_fail", {})
+        ap = val.get("adversarial_pass", {})
+        bn = val.get("benign", {})
+        console.print(f"  Validation:")
+        if af.get("total"):
+            console.print(f"    Failed adversarial caught: {af['caught']}/{af['total']} ({af.get('rate',0):.1%})")
+        if ap.get("total"):
+            console.print(f"    Passed adversarial caught: {ap['caught']}/{ap['total']} ({ap.get('rate',0):.1%})")
+        if bn.get("total"):
+            console.print(f"    Benign allowed: {bn['correct']}/{bn['total']} ({bn.get('rate',0):.1%})")
+
+
+
+
+# ---------------------------------------------------------------------------
+# Data fetching
+# ---------------------------------------------------------------------------
+
+def _fetch_experiments(client, category, exclude=False, last_n=10,
+                       from_date=None, until_date=None):
+    experiments = []
+    page = 1
+    while True:
+        result = client.list_experiments(page=page, size=50)
+        for exp in result.get("data", []):
+            if exp.get("status") != "Finished":
+                continue
+            has = category in exp.get("test_category", "")
+            if exclude and has:
+                continue
+            if not exclude and not has:
+                continue
+            created = exp.get("created_at", "")
+            if from_date and created < from_date:
+                continue
+            if until_date and created > until_date:
+                continue
+            experiments.append(exp)
+        if not result.get("has_next_page"):
+            break
+        page += 1
+    experiments.sort(key=lambda e: e.get("created_at", ""), reverse=True)
+    return experiments[:last_n]
+
+
+def _fetch_logs(client, experiments):
+    all_logs = []
+    with Progress(console=console) as progress:
+        task = progress.add_task("  Fetching...", total=len(experiments))
+        for exp in experiments:
+            cat = exp.get("test_category", "")
+            page = 1
+            while True:
+                result = client.get_experiment_logs(exp["id"], page=page, size=100)
+                for log in result.get("data", []):
+                    log["test_category"] = cat
+                all_logs.extend(result.get("data", []))
+                if not result.get("has_next_page"):
+                    break
+                page += 1
+            progress.advance(task)
+    return all_logs
+
+
+def _fetch_intents(client, project_id):
+    try:
+        data = client.get(f"projects/{project_id}")
+        intents = data.get("scope", {}).get("intents", {})
+        return intents.get("permitted", []), intents.get("restricted", [])
+    except Exception:
+        return None, None
+
+

{humanbound_cli-0.7.0 → humanbound_cli-0.9.0}/humanbound_cli/main.py

@@ -35,6 +35,7 @@ from .commands import (
     monitor,
     webhooks,
     assessments,
+    firewall,
     mcp,
 )
 
@@ -96,6 +97,7 @@ cli.add_command(webhooks.webhooks_group)
 cli.add_command(connectors.connectors_group)
 cli.add_command(inventory.inventory_group)
 cli.add_command(assessments.assessments_group)
+cli.add_command(firewall.firewall_group)
 cli.add_command(auth.auth_group)
 cli.add_command(orgs.orgs_group)
 cli.add_command(completion.completion_command)

{humanbound_cli-0.7.0 → humanbound_cli-0.9.0}/humanbound_cli.egg-info/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: humanbound-cli
-Version: 0.7.0
+Version: 0.9.0
 Summary: Humanbound CLI - command line interface for AI agent security testing.
 Author-email: Kostas Siabanis <hello@humanbound.ai>, Demetris Gerogiannis <hello@humanbound.ai>
 License: Apache-2.0
@@ -60,6 +60,7 @@ Humanbound runs automated adversarial attacks against your bot's live endpoint,
 | **Posture Scoring** | Quantified 0-100 security score with breakdown by findings, coverage, and resilience. Track over time. |
 | **Shadow AI Discovery** | Scan cloud tenants for AI services, assess risk with 15 SAI threat classes, and govern your AI inventory. |
 | **Guardrails Export** | Generate protection rules from test findings. Export to OpenAI or Humanbound format. |
+| **Firewall Training** | Train agent-specific Tier 2 classifiers from adversarial + QA test data. Pluggable model architecture via AgentClassifier scripts. |
 | **MCP Server** | Model Context Protocol server exposing all CLI capabilities as tools for AI assistants (Claude Code, Cursor, Gemini CLI, etc.). |
 
 ### Why Humanbound?
@@ -557,6 +558,30 @@ hb findings [--status open] [--severity high] [--json]
 hb guardrails [--vendor humanbound|openai] [--format json|yaml] [-o FILE]
 ```
 
+### Firewall
+
+Train agent-specific classifiers for [hb-firewall](https://github.com/humanbound/firewall).
+
+```bash
+# Train from adversarial + QA test data
+hb firewall train --model detectors/setfit_classifier.py
+
+# Show model info
+hb firewall show firewall.hbfw
+```
+
+| Flag | Default | Description |
+|------|---------|-------------|
+| `--model PATH` | — | Path to AgentClassifier script (required) |
+| `--last N` | 10 | Last N finished experiments |
+| `--from DATE` | — | Start date filter |
+| `--until DATE` | — | End date filter |
+| `--min-samples` | 30 | Minimum conversations required |
+| `--output` | firewall_\<project\>.hbfw | Output file path |
+| `--benign-dataset` | — | HuggingFace dataset for benign benchmarking |
+
+See [hb-firewall docs](https://github.com/humanbound/firewall) for the AgentClassifier interface and full integration guide.
+
 ### Shell Completion
 
 ```bash
@@ -723,6 +748,22 @@ hb connect -e ./bot-config.json
 hb test -e ./bot-config.json
 ```
 
+### Whitebox testing with telemetry
+
+Add a `telemetry` block to your agent config to enable whitebox testing. Humanbound fetches tool calls, memory operations, and resource usage from your observability platform (LangFuse, LangSmith, OpenAI Assistants, W&B, Helicone, AgentOps, or custom).
+
+```json
+{
+  "telemetry": {
+    "format": "langfuse",
+    "endpoint": "https://cloud.langfuse.com/api/public/sessions/$session_id",
+    "headers": { "Authorization": "Basic <base64(pk:sk)>" }
+  }
+}
+```
+
+See the full [Telemetry Integration Guide](https://docs.humanbound.ai/integrations/telemetry/) for vendor-specific setup and the custom extraction map reference.
+
 ### Shadow AI discovery & governance
 
 ```bash
@@ -759,6 +800,29 @@ claude mcp add humanbound -- hb mcp
 hb guardrails --vendor openai --format json -o guardrails.json
 ```
 
+### Train and deploy firewall
+
+```bash
+# Run adversarial tests
+hb test
+
+# Train Tier 1 classifier from results
+hb firewall train -o model.hbfw
+
+# Verify quality
+hb firewall show model.hbfw
+# F1=0.95, Precision=0.97, Tier 1 coverage=92%
+
+# Test before deploying
+# Deploy in your app
+python -c "
+from hb_firewall import Firewall
+fw = Firewall.from_config('agent.yaml', model_path='model.hbfw')
+result = fw.evaluate('What is my balance?')
+print(result.verdict, result.tier)  # Verdict.PASS 1
+"
+```
+
 ---
 
 ### On-Premises

{humanbound_cli-0.7.0 → humanbound_cli-0.9.0}/humanbound_cli.egg-info/SOURCES.txt

@@ -28,6 +28,7 @@ humanbound_cli/commands/discover.py
 humanbound_cli/commands/docs.py
 humanbound_cli/commands/experiments.py
 humanbound_cli/commands/findings.py
+humanbound_cli/commands/firewall.py
 humanbound_cli/commands/guardrails.py
 humanbound_cli/commands/init.py
 humanbound_cli/commands/inventory.py

{humanbound_cli-0.7.0 → humanbound_cli-0.9.0}/humanbound_cli.egg-info/top_level.txt

@@ -1,4 +1,3 @@
-build
 dist
 docs
 humanbound_cli

{humanbound_cli-0.7.0 → humanbound_cli-0.9.0}/pyproject.toml

@@ -4,7 +4,7 @@ build-backend = "setuptools.build_meta"
 
 [project]
 name = "humanbound-cli"
-version = "0.7.0"
+version = "0.9.0"
 authors = [
   { name="Kostas Siabanis", email="hello@humanbound.ai" },
   { name="Demetris Gerogiannis", email="hello@humanbound.ai" },