diff --git a/.gitignore b/.gitignore
index 8ac2a08..a6a443f 100644
--- a/.gitignore
+++ b/.gitignore
@@ -66,3 +66,5 @@ perfs_benchmark.log
 bench/
 .chromnn
 *.keras
+*.fa
+*.fa.fai
diff --git a/src/momics/cli/__init__.py b/src/momics/cli/__init__.py
index e7804da..19faecf 100644
--- a/src/momics/cli/__init__.py
+++ b/src/momics/cli/__init__.py
@@ -39,6 +39,8 @@ def cli(ctx):
     query,
     remove,
     tree,
+    consolidate,
+    manifest,
 )
 
 __all__ = [
@@ -52,4 +54,6 @@ def cli(ctx):
     "cp",
     "binnify",
     "cloudconfig",
+    "consolidate",
+    "manifest",
 ]
diff --git a/src/momics/cli/consolidate.py b/src/momics/cli/consolidate.py
new file mode 100644
index 0000000..dee0f8d
--- /dev/null
+++ b/src/momics/cli/consolidate.py
@@ -0,0 +1,37 @@
+import logging
+import click
+
+from momics import momics as m
+
+from . import cli
+
+
+@cli.command()
+@click.option(
+    "--vacuum",
+    "-v",
+    help="Flag to indicate wether fragments should be vacuumed after consolidation.",
+    is_flag=True,
+    required=False,
+)
+@click.argument("path", metavar="MOMICS_REPO", required=True)
+@click.pass_context
+def consolidate(ctx, path, vacuum):
+    """Consolidate a momics repository.
+
+    Consolidation is the process of compacting all the fragments from the repository
+    to remove any unused space. This process is useful to reduce the size of the
+    repository and improve performance.
+
+    The vacuum flag indicates wether the consolidated
+    array should be vacuumed after consolidation. This process is useful to further
+    reduce the size of the repository.
+    """
+
+    mom = m.Momics(path)
+    s = mom.size()
+    mom.consolidate(vacuum=vacuum)
+    sf = mom.size()
+    logging.info(f"Final repository size: {sf/(1024*1024)} Mb.")
+    logging.info(f"Space saved after consolidation: {(s - sf)/(1024*1024)} Mb.")
+    return True
diff --git a/src/momics/cli/cp.py b/src/momics/cli/cp.py
index ef100e3..40ec8ae 100644
--- a/src/momics/cli/cp.py
+++ b/src/momics/cli/cp.py
@@ -40,13 +40,14 @@
 def cp(ctx, path, type, label, force, output):
     """Copy sequence/track/feature set from a momics repo to a fa/bigwig/bed file."""
 
-    if not force:
-        click.confirm(
-            f"{output} file already exists. \
-            Are you sure you want to overwrite it",
-            abort=True,
-        )
-        os.remove(output)
+    if not force and output:
+        if os.path.exists(output):
+            click.confirm(
+                f"{output} file already exists. \
+                Are you sure you want to overwrite it",
+                abort=True,
+            )
+            os.remove(output)
 
     m = momics.Momics(path)
     if type == "sequence":
diff --git a/src/momics/cli/manifest.py b/src/momics/cli/manifest.py
new file mode 100644
index 0000000..502595a
--- /dev/null
+++ b/src/momics/cli/manifest.py
@@ -0,0 +1,48 @@
+import json
+import os
+import click
+
+from momics import momics as m
+
+from . import cli
+
+
+@cli.command()
+@click.option(
+    "--output",
+    "-o",
+    help="Path of output JSON file to write.",
+    type=str,
+    required=False,
+)
+@click.option(
+    "--force",
+    "-f",
+    help="Force overwrite of existing files.",
+    is_flag=True,
+    default=False,
+    show_default=True,
+)
+@click.argument("path", metavar="MOMICS_REPO", required=True)
+@click.pass_context
+def manifest(ctx, path, output, force):
+    """Print the manifest of a momics repository."""
+
+    if not force and output:
+        if os.path.exists(output):
+            click.confirm(
+                f"{output} file already exists. Are you sure you want to overwrite it",
+                abort=True,
+            )
+            os.remove(output)
+
+    mom = m.Momics(path)
+    man = mom.manifest()
+
+    if not output:
+        print(json.dumps(man, indent=2))
+        return None
+    else:
+        with open(output, "w") as f:
+            json.dump(man, f, indent=2)
+        return True
diff --git a/src/momics/momics.py b/src/momics/momics.py
index e6e18a9..3c60db2 100644
--- a/src/momics/momics.py
+++ b/src/momics/momics.py
@@ -1,3 +1,4 @@
+import collections
 import concurrent.futures
 import logging
 import os
@@ -130,12 +131,12 @@ def _create_chroms_schema(self, chr_lengths: dict) -> None:
             tiledb.Dim(
                 name="chrom_index",
                 domain=(0, len(chr_lengths) - 1),
-                dtype=np.int32,
+                dtype=np.uint32,
                 tile=len(chr_lengths),
             )
         )
         attr_chr = tiledb.Attr(name="chrom", dtype="ascii", var=True)
-        attr_length = tiledb.Attr(name="length", dtype=np.int64)
+        attr_length = tiledb.Attr(name="length", dtype=np.uint32)
         schema = tiledb.ArraySchema(
             ctx=self.cfg.ctx,
             domain=dom_genome,
@@ -154,7 +155,7 @@ def _create_sequence_schema(self, tile: int) -> None:
                 tiledb.Dim(
                     name="position",
                     domain=(0, chrom_length),
-                    dtype=np.int64,
+                    dtype=np.uint32,
                     tile=_set_tiledb_tile(tile, chrom_length),
                     filters=TILEDB_POSITION_FILTERS,
                 )
@@ -172,7 +173,7 @@ def _create_track_schema(self, max_bws: int, tile: int) -> None:
         # Create /coverage/tracks.tdb
         tdb = self._build_uri("coverage", "tracks.tdb")
         dom = tiledb.Domain(
-            tiledb.Dim(name="idx", domain=(0, max_bws), dtype=np.int64, tile=1),
+            tiledb.Dim(name="idx", domain=(0, max_bws), dtype=np.uint32, tile=1),
         )
         attr1 = tiledb.Attr(name="label", dtype="ascii")
         attr2 = tiledb.Attr(name="path", dtype="ascii")
@@ -188,7 +189,7 @@ def _create_track_schema(self, max_bws: int, tile: int) -> None:
                 tiledb.Dim(
                     name="position",
                     domain=(0, chrom_length),
-                    dtype=np.int64,
+                    dtype=np.uint32,
                     tile=_set_tiledb_tile(tile, chrom_length),
                     filters=TILEDB_POSITION_FILTERS,
                 )
@@ -234,14 +235,14 @@ def _create_features_schema(self, max_features: int, tile: int) -> None:
                 tiledb.Dim(
                     name="start",
                     domain=(0, chrom_length),
-                    dtype=np.int64,
+                    dtype=np.uint32,
                     tile=_set_tiledb_tile(tile, chrom_length),
                     filters=TILEDB_POSITION_FILTERS,
                 ),
                 tiledb.Dim(
                     name="stop",
                     domain=(0, chrom_length),
-                    dtype=np.int64,
+                    dtype=np.uint32,
                     tile=_set_tiledb_tile(tile, chrom_length),
                     filters=TILEDB_POSITION_FILTERS,
                 ),
@@ -948,3 +949,125 @@ def export_features(self, features: str, output: Path) -> "Momics":
         bed = self.features(features)
         bed.to_bed(output)
         return self
+
+    def manifest(self) -> dict:
+        """
+        Returns the manifest of the Momics repository. The manifest lists the
+        configuration for all the arrays stored in the repository, including
+        their schema, attributes, and metadata.
+        """
+
+        def is_tiledb_array(uri, ctx):
+            try:
+                return tiledb.object_type(uri, ctx) == "array"
+            except tiledb.TileDBError:
+                return False
+
+        def is_tiledb_group(uri, ctx):
+            try:
+                return tiledb.object_type(uri, ctx) == "group"
+            except tiledb.TileDBError:
+                return False
+
+        def get_array_schema(uri, ctx):
+            try:
+                with tiledb.open(uri, ctx=ctx, mode="r") as array:
+                    schema = array.schema
+
+                    d = collections.defaultdict(
+                        None,
+                        {
+                            "uri": uri,
+                            "type": "array",
+                            "shape": schema.shape,
+                            "sparse": schema.sparse,
+                            "chunksize": schema.capacity,
+                            "cell_order": schema.cell_order,
+                            "tile_order": schema.tile_order,
+                        },
+                    )
+
+                    d["dims"] = collections.defaultdict(None)
+                    for dim in schema.domain:
+                        d["dims"][dim.name] = {
+                            "domain": tuple([int(i) for i in dim.domain]),
+                            "dtype": str(dim.dtype),
+                            "tile": int(dim.tile),
+                            "filters": str(dim.filters),
+                        }
+                    d["dims"] = dict(d["dims"])
+
+                    d["attrs"] = collections.defaultdict(None)
+                    attrs = [schema.attr(x) for x in range(schema.nattr)]
+                    for attr in attrs:
+                        d["attrs"][attr.name] = {
+                            "dtype": str(attr.dtype),
+                            "filters": str(attr.filters),
+                        }
+                    d["attrs"] = dict(d["attrs"])
+
+                    d["modification_timestamps"] = {frag.uri: frag.timestamp_range for frag in tiledb.FragmentInfoList(uri)}
+                    d = dict(d)
+
+                return d
+
+            except Exception as e:
+                print(f"Error reading array schema from {uri}: {e}")
+                return None
+
+        def get_group_metadata(uri):
+            try:
+                return {"uri": uri, "type": "group"}
+            except Exception as e:
+                print(f"Error reading group metadata from {uri}: {e}")
+                return None
+
+        def traverse_directory(base_dir, ctx):
+            manifest = collections.defaultdict(None)
+            for pointer in self.cfg.vfs.ls_recursive(base_dir):
+                if is_tiledb_array(pointer, ctx):
+                    array_info = get_array_schema(pointer, ctx)
+                    if array_info:
+                        manifest[pointer] = array_info
+                elif is_tiledb_group(pointer, ctx):
+                    group_info = get_group_metadata(pointer)
+                    if group_info:
+                        manifest[pointer] = group_info
+
+            return manifest
+
+        man = traverse_directory(self.path, self.cfg.ctx)
+        return man
+
+    def consolidate(self, vacuum: bool = True) -> Literal[True]:
+        """
+        Consolidates the fragments of all arrays in the repository.
+
+        Args:
+            vacuum (bool, optional): Vacuum the consolidated array. Defaults to True.
+        """
+        for pointer in self.cfg.vfs.ls_recursive(self.path):
+            try:
+                if tiledb.object_type(pointer) == "array":
+                    logging.debug(f"Consolidating array at: {pointer}")
+                    tiledb.consolidate(pointer)
+                    if vacuum:
+                        tiledb.vacuum(pointer)
+            except tiledb.TileDBError as e:
+                print(f"Error processing {pointer}: {e}")
+
+        return True
+
+    def size(self) -> int:
+        """
+        Returns:
+            int: The size of the repository in bytes.
+        """
+        size = 0
+        for pointer in self.cfg.vfs.ls_recursive(self.path):
+            if self.cfg.vfs.is_file(pointer):
+                try:
+                    size += self.cfg.vfs.file_size(pointer)
+                except tiledb.TileDBError as e:
+                    print(f"Error processing {pointer}: {e}")
+        return size
diff --git a/tests/test_cli.py b/tests/test_cli.py
index 4cf1965..bde8c54 100644
--- a/tests/test_cli.py
+++ b/tests/test_cli.py
@@ -182,6 +182,22 @@ def test_config(runner):
     assert "aws_access_key_id: ABCD" in result.output
 
 
+def test_consolidate(runner, path):
+    m = Momics(path)
+    s = m.size()
+    result = runner.invoke(cli.consolidate.consolidate, ["--vacuum", path])
+    assert result.exit_code == 0
+    sf = m.size()
+    assert sf < s
+
+
+def test_manifest(runner, path):
+    result = runner.invoke(cli.manifest.manifest, ["--output", "out.json", "-f", path])
+    assert result.exit_code == 0
+    assert os.path.exists("out.json")
+    os.remove("out.json")
+
+
 def test_delete(runner, path):
     result = runner.invoke(cli.delete.delete, ["-y", "oiasudhncoaisuhmdcoiaushcd"])
     assert result.output == "Repository oiasudhncoaisuhmdcoiaushcd does not exist.\n"
diff --git a/tests/test_momics.py b/tests/test_momics.py
index 36c3b3b..a8970cb 100644
--- a/tests/test_momics.py
+++ b/tests/test_momics.py
@@ -1,3 +1,4 @@
+import collections
 from pathlib import Path
 import numpy as np
 import pyranges as pr
@@ -165,6 +166,22 @@ def test_Momics_binnify(momics_path: str):
     assert q.df.shape == (60, 3)
 
 
+@pytest.mark.order(2)
+def test_Momics_consolidate(momics_path: str):
+    mom = momics.Momics(momics_path)
+    s = mom.size()
+    x = mom.consolidate(vacuum=True)
+    assert x
+    assert mom.size() < s
+
+
+@pytest.mark.order(2)
+def test_Momics_manifest(momics_path: str):
+    mom = momics.Momics(momics_path)
+    man = mom.manifest()
+    assert isinstance(man, collections.defaultdict)
+
+
 @pytest.mark.order(2)
 def test_Momics_features(momics_path: str):
     mom = momics.Momics(momics_path)