diff --git a/docs/content/docs/proxy.mdx b/docs/content/docs/proxy.mdx
index 8747cd0a6..78a358348 100644
--- a/docs/content/docs/proxy.mdx
+++ b/docs/content/docs/proxy.mdx
@@ -37,7 +37,7 @@ Telemetry is **off by default** (opt-in). Opt in with `HEADROOM_TELEMETRY=on` or
 | `--max-connections` | `500` | Maximum upstream HTTP connections |
 | `--max-keepalive` | `100` | Maximum upstream keep-alive connections |
 | `--http-proxy` | None | HTTP proxy URL for upstream provider requests only; HTTPS provider APIs use CONNECT |
-| `--mode` | `token` | Optimization mode: `token` prioritizes compression, `cache` preserves provider prefix-cache stability |
+| `--mode` | `cache` | Optimization mode: `token` prioritizes compression, `cache` preserves provider prefix-cache stability. Default is `cache` (see [Savings profiles](#savings-profiles)) |
 | `--no-optimize` | `false` | Disable optimization (passthrough mode) |
 | `--no-cache` | `false` | Disable semantic caching |
 | `--no-rate-limit` | `false` | Disable rate limiting |
@@ -65,8 +65,8 @@ Avoid setting process-wide variables such as `HTTP_PROXY`, `HTTPS_PROXY`, `ALL_P
 
 | Option | Default | Description |
 |--------|---------|-------------|
-| `--mode token` | `token` | Prioritize token compression. This is the default. |
-| `--mode cache` | `token` | Preserve prior turns to maximize provider prefix-cache hit rate. |
+| `--mode token` | | Prioritize token compression; prior turns may be rewritten for maximum savings. |
+| `--mode cache` | default | Freeze prior turns to maximize provider prefix-cache hit rate. This is the effective default (see [Savings profiles](#savings-profiles)). |
 | `--intercept-tool-results` | `false` | Opt into tool-result interceptors such as ast-grep Read outlining. |
 | `--no-read-lifecycle` | `false` | Disable stale/superseded Read-output compression. |
 | `--code-aware` / `--no-code-aware` | disabled | Enable or disable AST-based code compression. Requires `headroom-ai[code]`. |
@@ -90,6 +90,26 @@ headroom proxy --mode token
 headroom proxy --mode cache
 ```
 
+### Savings profiles
+
+`HEADROOM_SAVINGS_PROFILE` selects a named profile that seeds Headroom's whole compression posture — proxy mode, keep-ratio, which messages are compressed, and `force_kompress` — at proxy startup. It is read by `headroom proxy` and by the `headroom wrap` subprocesses. When unset, the default profile is `coding`.
+
+| Profile | Target savings | Mode | Notes |
+|---------|----------------|------|-------|
+| `coding` | emergent (~50%) | `cache` | **Default.** Delta-only compression at ~0 prefix-cache busts; never lossy-compresses file reads. |
+| `balanced` | ~70% | `token` | Moderate compression with structural compaction. Also the fallback for an unknown profile name. |
+| `agent-90` | ~90% | `token` | Aggressive; pins a `0.10` keep-ratio and forces Kompress. |
+| `general` | emergent (~60%) | `token` | Non-coding workloads. |
+
+An unrecognized `HEADROOM_SAVINGS_PROFILE` value logs a warning and falls back to `balanced` — the proxy never fails to start over a bad profile name. See `headroom/agent_savings.py` for each profile's full set of knobs.
+
+Because the default `coding` profile uses **cache** mode (and the proxy's own default mode is also `cache`), Headroom runs in cache mode out of the box. Mode precedence: an explicit `--mode` wins, otherwise `HEADROOM_MODE` (which a profile seeds), otherwise the `cache` default. To run token mode, pass `--mode token` or choose a token-mode profile:
+
+```bash
+# Aggressive ~90% token-savings profile
+HEADROOM_SAVINGS_PROFILE=agent-90 headroom proxy --port 8787
+```
+
 ### Optional features
 
 | Option | Default | Description |
diff --git a/headroom/cli/proxy.py b/headroom/cli/proxy.py
index 0c60c5da4..c97cc5868 100644
--- a/headroom/cli/proxy.py
+++ b/headroom/cli/proxy.py
@@ -247,7 +247,7 @@ def dashboard(port: int, no_open: bool) -> None:
         case_sensitive=False,
     ),
     help=(
-        "Optimization mode (default: token).\n"
+        "Optimization mode (default: cache).\n"
         "  token  — prioritize compression; prior turns may be rewritten for max savings.\n"
         "  cache  — freeze prior turns to maximise provider prefix-cache hit rate.\n"
         "Legacy aliases (token_mode, token_savings, token_headroom, cache_mode, "
diff --git a/tests/test_cli_proxy_improvements.py b/tests/test_cli_proxy_improvements.py
index d6787da13..e400ae35b 100644
--- a/tests/test_cli_proxy_improvements.py
+++ b/tests/test_cli_proxy_improvements.py
@@ -478,6 +478,11 @@ class TestHelpTextCompleteness:
     def test_help_contains_mode_option(self, runner: CliRunner) -> None:
         assert "--mode" in self._help(runner)
 
+    def test_help_reports_cache_as_default_mode(self, runner: CliRunner) -> None:
+        out = self._help(runner)
+        assert "Optimization mode (default: cache)" in out
+        assert "Optimization mode (default: token)" not in out
+
     def test_help_contains_workers_option(self, runner: CliRunner) -> None:
         assert "--workers" in self._help(runner)