Merge pull request 'git-gate: ExtraHosts on bottle.git entries' (#12) from git-gate-extra-hosts into main

README.md

@@ -118,8 +118,14 @@ that enforces the manifest before it leaves the host.
   upstream has *now* (fail-closed if unreachable). The agent's
   `~/.gitconfig` rewrites the real URL to the gate via `insteadOf`,
   so push, fetch, clone, and pull all route through. The agent
-  never sees the upstream credential. Brought up only when
-  `bottle.git` has entries. Design in `docs/prds/0008-git-gate.md`.
+  never sees the upstream credential. If the upstream's hostname
+  isn't resolvable from the gate container (e.g. a Tailscale-only
+  host whose public DNS points elsewhere), pin its IP via
+  `ExtraHosts: { "<hostname>": "<ip>" }` on the `bottle.git` entry —
+  the gate's `/etc/hosts` gets the override while the agent's
+  `insteadOf` rewrite still keys off the original hostname. Brought
+  up only when `bottle.git` has entries. Design in
+  `docs/prds/0008-git-gate.md`.
 
 When the agent exits, `cli.py` tears down every sidecar that was
 brought up and the two networks; nothing about a bottle persists

claude-bottle.example.json

@@ -19,14 +19,13 @@
         "GIT_AUTHOR_NAME": "Eric Diderich",
         "NODE_ENV":        "development"
       },
-      "ssh": [
+      "git": [
         {
-          "Host":         "gitea",
-          "Hostname":     "gitea.dideric.is",
-          "User":         "git",
-          "Port":         30009,
+          "Name":         "claude-bottle",
+          "Upstream":     "ssh://git@gitea.dideric.is:30009/didericis/claude-bottle.git",
           "IdentityFile": "/Users/didericis/.ssh/id_ed25519_gitea",
-          "KnownHostKey": "gitea.dideric.is ssh-ed25519 AAAA..."
+          "KnownHostKey": "ssh-ed25519 AAAA...",
+          "ExtraHosts":   { "gitea.dideric.is": "100.78.141.42" }
         }
       ],
       "egress": {

claude_bottle/backend/docker/git_gate.py

@@ -8,7 +8,12 @@ import os
 import subprocess
 from pathlib import Path
 
-from ...git_gate import GitGate, GitGatePlan, git_gate_known_hosts_line
+from ...git_gate import (
+    GitGate,
+    GitGatePlan,
+    git_gate_aggregate_extra_hosts,
+    git_gate_known_hosts_line,
+)
 from ...log import die, info, warn
 from ...util import expand_tilde
 from . import util as docker_mod
@@ -101,8 +106,10 @@ class DockerGitGate(GitGate):
             "docker", "create",
             "--name", name,
             "--network", plan.internal_network,
-            GIT_GATE_IMAGE,
         ]
+        for host, ip in git_gate_aggregate_extra_hosts(plan.upstreams).items():
+            create_args.extend(["--add-host", f"{host}:{ip}"])
+        create_args.append(GIT_GATE_IMAGE)
         if subprocess.run(
             create_args,
             stdout=subprocess.DEVNULL,

claude_bottle/git_gate.py

@@ -30,12 +30,18 @@ backend-specific and lives on concrete subclasses (see
 from __future__ import annotations
 
 from abc import ABC, abstractmethod
-from dataclasses import dataclass
+from dataclasses import dataclass, field
 from pathlib import Path
+from typing import Mapping
 
+from .log import die
 from .manifest import Bottle
 
 
+def _empty_str_map() -> dict[str, str]:
+    return {}
+
+
 @dataclass(frozen=True)
 class GitGateUpstream:
     """One bare repo on the gate. `name` drives the bare-repo path
@@ -47,7 +53,12 @@ class GitGateUpstream:
     `identity_file` is the host-side absolute path the gate's start
     step will docker-cp into the container. `known_host_key` is the
     KnownHostKey string from the manifest; the gate's start step
-    materialises it into a known_hosts file if non-empty."""
+    materialises it into a known_hosts file if non-empty.
+
+    `extra_hosts` is a `{hostname: ip}` map the backend injects into
+    the gate container's `/etc/hosts` via `--add-host` so the gate
+    can resolve upstream hostnames that aren't reachable via the
+    container's default DNS (e.g. Tailscale-only hosts)."""
 
     name: str
     upstream_url: str
@@ -55,6 +66,7 @@ class GitGateUpstream:
     upstream_port: str
     identity_file: str
     known_host_key: str
+    extra_hosts: Mapping[str, str] = field(default_factory=_empty_str_map)
 
 
 @dataclass(frozen=True)
@@ -92,11 +104,38 @@ def git_gate_upstreams_for_bottle(bottle: Bottle) -> tuple[GitGateUpstream, ...]
             upstream_port=e.UpstreamPort,
             identity_file=e.IdentityFile,
             known_host_key=e.KnownHostKey,
+            extra_hosts=dict(e.ExtraHosts),
         )
         for e in bottle.git
     )
 
 
+def git_gate_aggregate_extra_hosts(
+    upstreams: tuple[GitGateUpstream, ...],
+) -> dict[str, str]:
+    """Merge every upstream's `extra_hosts` into a single
+    `{hostname: ip}` map for `--add-host` on the gate container. Two
+    entries naming the same hostname with different IPs is a manifest
+    bug — the gate has one /etc/hosts — so die loudly with the
+    conflicting names rather than silently picking one."""
+    merged: dict[str, str] = {}
+    source: dict[str, str] = {}
+    for u in upstreams:
+        for host, ip in u.extra_hosts.items():
+            existing = merged.get(host)
+            if existing is None:
+                merged[host] = ip
+                source[host] = u.name
+            elif existing != ip:
+                die(
+                    f"git-gate ExtraHosts conflict: '{host}' maps to "
+                    f"'{existing}' in upstream '{source[host]}' and to "
+                    f"'{ip}' in upstream '{u.name}'. The gate has one "
+                    f"/etc/hosts; pick one IP."
+                )
+    return merged
+
+
 def git_gate_known_hosts_line(host: str, port: str, key: str) -> str:
     """Format `host[:port] key` for OpenSSH's known_hosts. Non-default
     ports use the bracketed `[host]:port` form (the form OpenSSH writes

claude_bottle/manifest.py

@@ -89,6 +89,13 @@ class GitEntry:
     upstream after gitleaks passes. The agent itself never holds the
     upstream credential.
 
+    `ExtraHosts` is an optional `{hostname: ip}` map injected into the
+    gate container's `/etc/hosts` via `--add-host`. Use it when the
+    Upstream's hostname isn't resolvable from the gate (e.g. a
+    Tailscale-only host whose public DNS A record points elsewhere):
+    the agent's `insteadOf` rewrite still matches the original
+    hostname, but the gate routes to the right IP.
+
     The Upstream URL is parsed once at construction and the pieces are
     stashed in the `Upstream*` fields so the git-gate render step
     doesn't have to re-parse."""
@@ -97,6 +104,7 @@ class GitEntry:
     Upstream: str
     IdentityFile: str
     KnownHostKey: str = ""
+    ExtraHosts: Mapping[str, str] = field(default_factory=_empty_str_dict)
     UpstreamUser: str = ""
     UpstreamHost: str = ""
     UpstreamPort: str = ""
@@ -124,6 +132,9 @@ class GitEntry:
             d.get("KnownHostKey"),
             f"bottle '{bottle_name}' git '{name}' KnownHostKey",
         )
+        extra_hosts = _opt_extra_hosts(
+            d.get("ExtraHosts"), f"bottle '{bottle_name}' git '{name}' ExtraHosts"
+        )
         user, host, port, path = _parse_git_upstream(
             upstream, f"bottle '{bottle_name}' git '{name}' Upstream"
         )
@@ -132,6 +143,7 @@ class GitEntry:
             Upstream=upstream,
             IdentityFile=ident,
             KnownHostKey=khk,
+            ExtraHosts=extra_hosts,
             UpstreamUser=user,
             UpstreamHost=host,
             UpstreamPort=port,
@@ -435,6 +447,26 @@ def _opt_port(value: object, label: str) -> str:
     die(f"{label} must be a string or number (was {type(value).__name__})")
 
 
+def _opt_extra_hosts(value: object, label: str) -> dict[str, str]:
+    """Validate a `{hostname: ip}` object and return a plain dict. None
+    yields an empty dict so callers can treat ExtraHosts as always
+    present. IP format is not checked here; docker validates at
+    `--add-host` time."""
+    if value is None:
+        return {}
+    obj = _as_json_object(value, label)
+    out: dict[str, str] = {}
+    for host, ip in obj.items():
+        if not host:
+            die(f"{label} contains an empty hostname key")
+        if not isinstance(ip, str):
+            die(f"{label}['{host}'] must be a string (was {type(ip).__name__})")
+        if not ip:
+            die(f"{label}['{host}'] must be a non-empty string")
+        out[host] = ip
+    return out
+
+
 def _parse_git_upstream(url: str, label: str) -> tuple[str, str, str, str]:
     """Parse `ssh://user@host[:port]/path` into (user, host, port, path).
     Dies if `url` doesn't match the ssh:// shape v1 supports. Default

docs/prds/0008-git-gate.md

@@ -83,7 +83,12 @@ for a declared upstream:
 - **Manifest field.** `bottle.git` — a list of git remotes the
   bottle is allowed to talk to, each with the credential the gate
   uses to push upstream. The agent gets no parallel `bottle.ssh`
-  entry for those upstreams.
+  entry for those upstreams. Each entry may also carry an
+  `ExtraHosts: { hostname: ip }` map, surfaced to the gate as
+  `--add-host` so the gate can resolve upstreams whose public DNS
+  doesn't point at the reachable IP (e.g. Tailscale-only hosts).
+  The agent-side `insteadOf` rewrite keys off the original hostname,
+  so the manifest's `Upstream` URL stays human-readable.
 - **Agent-side URL rewrite.** Provisioner emits `~/.gitconfig`
   with `[url "<gate-url>"] insteadOf = <real-url>` so every git
   operation against the declared upstream (push, fetch, clone,

tests/unit/test_git_gate.py

@@ -9,12 +9,15 @@ from claude_bottle.git_gate import (
     GitGate,
     GitGatePlan,
     GitGateUpstream,
+    git_gate_aggregate_extra_hosts,
     git_gate_known_hosts_line,
     git_gate_render_access_hook,
     git_gate_render_entrypoint,
     git_gate_render_hook,
     git_gate_upstreams_for_bottle,
 )
+from claude_bottle.log import Die
+from claude_bottle.manifest import Manifest
 from tests.fixtures import fixture_minimal, fixture_with_git
 
 
@@ -43,6 +46,84 @@ class TestUpstreamsForBottle(unittest.TestCase):
         self.assertEqual((), git_gate_upstreams_for_bottle(bottle))
 
 
+class TestExtraHostsPlumbing(unittest.TestCase):
+    def test_upstream_carries_extra_hosts_from_manifest(self):
+        m = Manifest.from_json_obj({
+            "bottles": {
+                "dev": {
+                    "git": [{
+                        "Name": "claude-bottle",
+                        "Upstream": "ssh://git@gitea.dideric.is:30009/didericis/claude-bottle.git",
+                        "IdentityFile": "/dev/null",
+                        "ExtraHosts": {"gitea.dideric.is": "100.78.141.42"},
+                    }],
+                },
+            },
+            "agents": {"demo": {"skills": [], "prompt": "", "bottle": "dev"}},
+        })
+        ups = git_gate_upstreams_for_bottle(m.bottles["dev"])
+        self.assertEqual(
+            {"gitea.dideric.is": "100.78.141.42"}, dict(ups[0].extra_hosts)
+        )
+
+    def test_aggregator_merges_distinct_hostnames(self):
+        ups = (
+            GitGateUpstream(
+                name="a", upstream_url="", upstream_host="", upstream_port="",
+                identity_file="", known_host_key="",
+                extra_hosts={"a.example": "10.0.0.1"},
+            ),
+            GitGateUpstream(
+                name="b", upstream_url="", upstream_host="", upstream_port="",
+                identity_file="", known_host_key="",
+                extra_hosts={"b.example": "10.0.0.2"},
+            ),
+        )
+        self.assertEqual(
+            {"a.example": "10.0.0.1", "b.example": "10.0.0.2"},
+            git_gate_aggregate_extra_hosts(ups),
+        )
+
+    def test_aggregator_allows_same_host_same_ip(self):
+        # Two entries listing the same host:ip is harmless duplication,
+        # not a conflict. The gate's /etc/hosts ends up with one line.
+        ups = (
+            GitGateUpstream(
+                name="a", upstream_url="", upstream_host="", upstream_port="",
+                identity_file="", known_host_key="",
+                extra_hosts={"gitea.dideric.is": "100.78.141.42"},
+            ),
+            GitGateUpstream(
+                name="b", upstream_url="", upstream_host="", upstream_port="",
+                identity_file="", known_host_key="",
+                extra_hosts={"gitea.dideric.is": "100.78.141.42"},
+            ),
+        )
+        self.assertEqual(
+            {"gitea.dideric.is": "100.78.141.42"},
+            git_gate_aggregate_extra_hosts(ups),
+        )
+
+    def test_aggregator_rejects_conflicting_ips(self):
+        ups = (
+            GitGateUpstream(
+                name="a", upstream_url="", upstream_host="", upstream_port="",
+                identity_file="", known_host_key="",
+                extra_hosts={"gitea.dideric.is": "100.78.141.42"},
+            ),
+            GitGateUpstream(
+                name="b", upstream_url="", upstream_host="", upstream_port="",
+                identity_file="", known_host_key="",
+                extra_hosts={"gitea.dideric.is": "10.0.0.99"},
+            ),
+        )
+        with self.assertRaises(Die):
+            git_gate_aggregate_extra_hosts(ups)
+
+    def test_aggregator_empty_is_empty(self):
+        self.assertEqual({}, git_gate_aggregate_extra_hosts(()))
+
+
 class TestKnownHostsLine(unittest.TestCase):
     def test_default_port_unbracketed(self):
         line = git_gate_known_hosts_line("github.com", "22", "ssh-ed25519 AAAA")

tests/unit/test_manifest_git.py

@@ -111,6 +111,53 @@ class TestGitEntryParsing(unittest.TestCase):
             }]))
 
 
+class TestGitEntryExtraHosts(unittest.TestCase):
+    def test_extra_hosts_defaults_to_empty(self):
+        m = Manifest.from_json_obj(_manifest([{
+            "Name": "foo",
+            "Upstream": "ssh://git@github.com/foo.git",
+            "IdentityFile": "/dev/null",
+        }]))
+        self.assertEqual({}, dict(m.bottles["dev"].git[0].ExtraHosts))
+
+    def test_extra_hosts_parses_host_to_ip_map(self):
+        m = Manifest.from_json_obj(_manifest([{
+            "Name": "claude-bottle",
+            "Upstream": "ssh://git@gitea.dideric.is:30009/didericis/claude-bottle.git",
+            "IdentityFile": "/dev/null",
+            "ExtraHosts": {"gitea.dideric.is": "100.78.141.42"},
+        }]))
+        eh = dict(m.bottles["dev"].git[0].ExtraHosts)
+        self.assertEqual({"gitea.dideric.is": "100.78.141.42"}, eh)
+
+    def test_extra_hosts_must_be_object(self):
+        with self.assertRaises(Die):
+            Manifest.from_json_obj(_manifest([{
+                "Name": "foo",
+                "Upstream": "ssh://git@github.com/foo.git",
+                "IdentityFile": "/dev/null",
+                "ExtraHosts": ["gitea.dideric.is", "100.78.141.42"],
+            }]))
+
+    def test_extra_hosts_ip_must_be_string(self):
+        with self.assertRaises(Die):
+            Manifest.from_json_obj(_manifest([{
+                "Name": "foo",
+                "Upstream": "ssh://git@github.com/foo.git",
+                "IdentityFile": "/dev/null",
+                "ExtraHosts": {"gitea.dideric.is": 100},
+            }]))
+
+    def test_extra_hosts_empty_ip_dies(self):
+        with self.assertRaises(Die):
+            Manifest.from_json_obj(_manifest([{
+                "Name": "foo",
+                "Upstream": "ssh://git@github.com/foo.git",
+                "IdentityFile": "/dev/null",
+                "ExtraHosts": {"gitea.dideric.is": ""},
+            }]))
+
+
 class TestGitEntryCrossValidation(unittest.TestCase):
     def test_duplicate_name_dies(self):
         with self.assertRaises(Die):