feat. Review foot note, docker fix, pass message to reviewer , update tests

AGENTS.md

@@ -100,7 +100,7 @@ pytest
 Docker compose:
 
 ```bash
-docker compose up --build
+docker compose up --build -f docker-compose.dev.yml
 ```
 
 ## Environment Contract
@@ -171,3 +171,9 @@ Treat these as high-sensitivity areas when modifying worker/runner paths.
 3. Add/update tests with behavior changes.
 4. Run `pytest`.
 5. Summarize impact, risks, and follow-ups in PR/commit notes.
+
+## Commiting After Completion
+If you are confident that your changes are ready to be committed, please follow the commit message format below:
+
+```[TYPE] Short description (max 50 chars)```
+Push after commiting. Ask the user once if you have permission to commit and from then on commit without asking.

TODO.md

@@ -24,7 +24,7 @@
 - [ ] `FEATURE`: Add username as possible command prefix, ex. "@bot-name review" in addition to "@codex review", for better UX discoverability.
 
 ### P2 (Nice to Have)
-- [ ] `FEATURE`: Add a note line generated by the reviewer at the end of comments to show model tokens used and such.
+- [x] `FEATURE`: Add a note line at the end of comments to show model tokens used and such.
 - [ ] `FEATURE`: Little static tailwind cdn styled page for any http endpoint that just shows what this is, incase this gets discovered by some random lad. Other routes than "/" should return a 404 with if a browser accessed it a again, tailwind cdn themed 404 page. Both should be nicely designed and minimalistic.
 - [ ] `FEATURE`: Apply `.codex-review.yml` `review.default_mode` when `@codex review` is issued without explicit mode.
 - [ ] `FEATURE`: Add per-repo command policy in `.codex-review.yml` for enabling/disabling `review`, `fix`, `explain`, and `rerun` independently.

alembic/versions/0002_review_job_trigger_comment_body.py

@@ -0,0 +1,25 @@
+"""add trigger comment body to review jobs
+
+Revision ID: 0002_review_job_trigger_comment_body
+Revises: 0001_initial
+Create Date: 2026-05-22 20:15:00
+"""
+
+from typing import Sequence, Union
+
+import sqlalchemy as sa
+from alembic import op
+
+
+revision: str = "0002_review_job_trigger_comment_body"
+down_revision: Union[str, None] = "0001_initial"
+branch_labels: Union[str, Sequence[str], None] = None
+depends_on: Union[str, Sequence[str], None] = None
+
+
+def upgrade() -> None:
+    op.add_column("review_jobs", sa.Column("trigger_comment_body", sa.Text(), nullable=True))
+
+
+def downgrade() -> None:
+    op.drop_column("review_jobs", "trigger_comment_body")

docker-compose.dev.yml

@@ -19,9 +19,6 @@ services:
 
   bot:
     build: .
-    depends_on:
-      mariadb:
-        condition: service_healthy
     env_file:
       - .env
     environment:

src/gitea_codex_bot/main.py

@@ -219,6 +219,7 @@ async def gitea_webhook(
             pr_number=pr_number,
             head_sha=head_sha,
             trigger_comment_id=comment_id,
+            trigger_comment_body=comment_body,
             requested_by=sender_username,
             command=parsed_command,
         )
@@ -232,6 +233,7 @@ async def gitea_webhook(
             pr_number=pr_number,
             head_sha=head_sha,
             trigger_comment_id=comment_id,
+            trigger_comment_body=comment_body,
             requested_by=sender_username,
             command=parsed_command,
         )

src/gitea_codex_bot/models.py

@@ -49,6 +49,7 @@ class ReviewJob(Base):
     pr_number: Mapped[int] = mapped_column(Integer, nullable=False)
     head_sha: Mapped[str] = mapped_column(String(64), nullable=False)
     trigger_comment_id: Mapped[int] = mapped_column(Integer, nullable=False)
+    trigger_comment_body: Mapped[str | None] = mapped_column(Text, nullable=True)
     command: Mapped[str] = mapped_column(String(64), nullable=False, default="review")
     command_args: Mapped[str | None] = mapped_column(Text, nullable=True)
     requested_by: Mapped[str] = mapped_column(String(255), nullable=False)

src/gitea_codex_bot/services/jobs.py

@@ -61,6 +61,7 @@ def enqueue_job(
     pr_number: int,
     head_sha: str,
     trigger_comment_id: int,
+    trigger_comment_body: str | None,
     requested_by: str,
     command: ParsedCommand,
 ) -> ReviewJob:
@@ -69,6 +70,7 @@ def enqueue_job(
         pr_number=pr_number,
         head_sha=head_sha,
         trigger_comment_id=trigger_comment_id,
+        trigger_comment_body=trigger_comment_body,
         command=command.name,
         command_args=" ".join(command.arguments) if command.arguments else None,
         requested_by=requested_by,

src/gitea_codex_bot/services/review_format.py

@@ -34,9 +34,13 @@ def format_unsupported_ack(command: ParsedCommand) -> str:
 
 
 def format_result_comment(head_sha: str, result: dict) -> str:
+    usage_note = _format_usage_note(result)
     markdown_comment = result.get("markdown_comment")
     if isinstance(markdown_comment, str) and markdown_comment.strip():
-        return _inject_head_sha_marker(head_sha, markdown_comment)
+        body = markdown_comment.strip()
+        if usage_note:
+            body = f"{body}\n\n{usage_note}"
+        return _inject_head_sha_marker(head_sha, body)
 
     verdict = result.get("verdict", "has_issues")
     confidence = float(result.get("confidence", 0.0))
@@ -64,4 +68,32 @@ def format_result_comment(head_sha: str, result: dict) -> str:
                     f"   Suggestion: {suggestion}" if suggestion else "   Suggestion: n/a",
                 ]
             )
-    return _inject_head_sha_marker(head_sha, "\n".join(lines).strip())
+    body = "\n".join(lines).strip()
+    if usage_note:
+        body = f"{body}\n\n{usage_note}"
+    return _inject_head_sha_marker(head_sha, body)
+
+
+def _format_usage_note(result: dict) -> str:
+    meta = result.get("_meta")
+    if not isinstance(meta, dict):
+        return ""
+
+    model = meta.get("model")
+    model_text = model.strip() if isinstance(model, str) and model.strip() else "unknown"
+
+    usage = meta.get("usage")
+    if not isinstance(usage, dict):
+        return f"_Note: model `{model_text}`._"
+
+    input_tokens = usage.get("input_tokens")
+    output_tokens = usage.get("output_tokens")
+    total_tokens = usage.get("total_tokens")
+    parts = [f"model `{model_text}`"]
+    if isinstance(input_tokens, int):
+        parts.append(f"input `{input_tokens}`")
+    if isinstance(output_tokens, int):
+        parts.append(f"output `{output_tokens}`")
+    if isinstance(total_tokens, int):
+        parts.append(f"total `{total_tokens}`")
+    return f"_Note: {', '.join(parts)} tokens used._"

src/gitea_codex_bot/services/reviewer.py

@@ -129,6 +129,7 @@ def _build_prompt(
         "}\n\n"
         f"PR URL: {pr.html_url}\n"
         f"Mode: {mode}\n"
+        f"Trigger message: {command.raw}\n"
         f"Repo focus: {', '.join(repo_cfg.focus)}\n"
         f"Diff truncated: {diff_context['truncated']}\n"
         f"Changed files:\n{os.linesep.join(diff_context['changed_files'])}\n\n"
@@ -166,10 +167,31 @@ def _call_openai_review(settings: Settings, prompt: str) -> dict[str, Any]:
         for content in item.get("content", []):
             text_value = content.get("text")
             if text_value:
-                return json.loads(text_value)
+                result = json.loads(text_value)
+                if isinstance(result, dict):
+                    result["_meta"] = _build_openai_result_meta(payload, settings)
+                return result
     raise ReviewError("OpenAI response did not contain JSON output text.")
 
 
+def _build_openai_result_meta(payload: dict[str, Any], settings: Settings) -> dict[str, Any]:
+    usage_raw = payload.get("usage")
+    usage: dict[str, int] = {}
+    if isinstance(usage_raw, dict):
+        for output_key, source_key in (
+            ("input_tokens", "input_tokens"),
+            ("output_tokens", "output_tokens"),
+            ("total_tokens", "total_tokens"),
+        ):
+            value = usage_raw.get(source_key)
+            if isinstance(value, int):
+                usage[output_key] = value
+    model = payload.get("model")
+    if not isinstance(model, str) or not model.strip():
+        model = settings.openai_review_model
+    return {"source": "openai_api", "model": model, "usage": usage}
+
+
 def _summarize_openai_failure(exc: Exception) -> str:
     if isinstance(exc, httpx.HTTPStatusError):
         status = exc.response.status_code

src/gitea_codex_bot/workers/container_runner.py

@@ -48,6 +48,7 @@ def run_review_ephemeral(
         if completed.returncode != 0:
             raise RuntimeError(_format_runner_failure(completed))
         parsed = _parse_codex_exec_stdout(completed.stdout)
+        parsed["_meta"] = _extract_result_meta_from_codex_stdout(completed.stdout, settings)
         return normalize_review_result(parsed)
     except Exception as exc:
         if settings.codex_auth_mode == "chatgpt":
@@ -202,6 +203,67 @@ def _parse_codex_exec_stdout(stdout: str) -> dict[str, Any]:
     raise RuntimeError(f"codex exec output text did not contain review JSON; text_tail={_tail_text(last_text, 400)}")
 
 
+def _extract_result_meta_from_codex_stdout(stdout: str, settings: Settings) -> dict[str, Any]:
+    model = settings.openai_review_model
+    usage: dict[str, int] = {}
+    for line in stdout.splitlines():
+        line = line.strip()
+        if not line:
+            continue
+        try:
+            payload = json.loads(line)
+        except json.JSONDecodeError:
+            continue
+        discovered_model = _find_first_string_for_key(payload, "model")
+        if discovered_model:
+            model = discovered_model
+        discovered_usage = _find_first_dict_for_key(payload, "usage")
+        if isinstance(discovered_usage, dict):
+            for output_key, source_key in (
+                ("input_tokens", "input_tokens"),
+                ("output_tokens", "output_tokens"),
+                ("total_tokens", "total_tokens"),
+            ):
+                value = discovered_usage.get(source_key)
+                if isinstance(value, int):
+                    usage[output_key] = value
+    return {"source": "ephemeral_runner", "model": model, "usage": usage}
+
+
+def _find_first_string_for_key(payload: Any, key: str) -> str | None:
+    if isinstance(payload, dict):
+        value = payload.get(key)
+        if isinstance(value, str) and value.strip():
+            return value
+        for nested in payload.values():
+            found = _find_first_string_for_key(nested, key)
+            if found:
+                return found
+    if isinstance(payload, list):
+        for item in payload:
+            found = _find_first_string_for_key(item, key)
+            if found:
+                return found
+    return None
+
+
+def _find_first_dict_for_key(payload: Any, key: str) -> dict[str, Any] | None:
+    if isinstance(payload, dict):
+        value = payload.get(key)
+        if isinstance(value, dict):
+            return value
+        for nested in payload.values():
+            found = _find_first_dict_for_key(nested, key)
+            if found:
+                return found
+    if isinstance(payload, list):
+        for item in payload:
+            found = _find_first_dict_for_key(item, key)
+            if found:
+                return found
+    return None
+
+
 def _parse_review_json_from_text(text: str) -> dict[str, Any] | None:
     candidates: list[str] = [text.strip()]
     fenced = re.search(r"```(?:json)?\s*(\{.*\})\s*```", text, flags=re.DOTALL | re.IGNORECASE)

src/gitea_codex_bot/workers/dispatcher.py

@@ -24,7 +24,8 @@ logger = logging.getLogger(__name__)
 
 def _command_from_job(job: ReviewJob) -> ParsedCommand:
     args = job.command_args.split() if job.command_args else []
-    return ParsedCommand(name=job.command, raw=f"@codex {job.command}", arguments=args, full="--full" in args, branch_fix="--branch" in args)
+    raw = (job.trigger_comment_body or "").strip() or f"@codex {job.command}"
+    return ParsedCommand(name=job.command, raw=raw, arguments=args, full="--full" in args, branch_fix="--branch" in args)
 
 
 def _handle_non_review_command(

tests/test_container_runner.py

@@ -9,6 +9,7 @@ from gitea_codex_bot.workers.container_runner import (
     CONTAINER_CODEX_HOME,
     _build_docker_command,
     _build_install_and_run_command,
+    _extract_result_meta_from_codex_stdout,
     _load_codex_auth_json_b64,
     _parse_codex_exec_stdout,
     _resolve_codex_auth_json_path,
@@ -145,3 +146,18 @@ def test_parse_codex_exec_stdout_from_fenced_json_text() -> None:
     parsed = _parse_codex_exec_stdout(stdout)
     assert parsed["verdict"] == "has_issues"
     assert parsed["summary"] == "x"
+
+
+def test_extract_result_meta_from_codex_stdout_collects_model_and_usage() -> None:
+    settings = get_settings()
+    stdout = '\n'.join(
+        [
+            '{"type":"response.started","model":"gpt-5.3-codex"}',
+            '{"type":"response.completed","response":{"usage":{"input_tokens":101,"output_tokens":22,"total_tokens":123}}}',
+        ]
+    )
+    meta = _extract_result_meta_from_codex_stdout(stdout, settings)
+    assert meta["model"] == "gpt-5.3-codex"
+    assert meta["usage"]["input_tokens"] == 101
+    assert meta["usage"]["output_tokens"] == 22
+    assert meta["usage"]["total_tokens"] == 123

tests/test_dispatcher.py

@@ -23,6 +23,7 @@ def test_process_one_job_recreates_persistent_comment_when_edit_returns_404(monk
             pr_number=9,
             head_sha="deadbeef",
             trigger_comment_id=111,
+            trigger_comment_body="@codex review",
             requested_by="alice",
             command=ParsedCommand(name="review", raw="@codex review"),
         )
@@ -71,3 +72,44 @@ def test_process_one_job_recreates_persistent_comment_when_edit_returns_404(monk
         assert persisted_comment_id == 990
         stored_job = session.execute(select(ReviewJob).where(ReviewJob.id == job.id)).scalar_one()
         assert stored_job.status.value == "succeeded"
+
+
+def test_process_one_job_passes_full_trigger_message_to_runner(monkeypatch) -> None:
+    captured: dict[str, str] = {}
+    session_factory = get_session_factory()
+    with session_factory() as session:
+        enqueue_job(
+            session,
+            repo="acme/repo",
+            pr_number=10,
+            head_sha="cafebabe",
+            trigger_comment_id=112,
+            trigger_comment_body="@codex review security --full\nFocus auth/session handling.",
+            requested_by="alice",
+            command=ParsedCommand(name="review", raw="@codex review security --full", arguments=["security", "--full"]),
+        )
+
+    def _fake_run_review_ephemeral(_settings, *, repo: str, pr_number: int, command: ParsedCommand):
+        captured["raw"] = command.raw
+        return {"verdict": "correct", "confidence": 0.9, "summary": "ok", "findings": []}
+
+    class _FakeGiteaClient:
+        def __init__(self, _settings) -> None:
+            pass
+
+        def get_pull_request(self, _repo: str, _pr_number: int):
+            return SimpleNamespace(is_fork=False)
+
+        def post_issue_comment(self, _repo: str, _pr_number: int, _body: str) -> int:
+            return 901
+
+        def edit_issue_comment(self, _repo: str, _comment_id: int, _body: str) -> int:
+            return _comment_id
+
+    monkeypatch.setattr("gitea_codex_bot.workers.dispatcher.run_review_ephemeral", _fake_run_review_ephemeral)
+    monkeypatch.setattr("gitea_codex_bot.workers.dispatcher.GiteaClient", _FakeGiteaClient)
+
+    settings = get_settings()
+    processed = process_one_job(settings)
+    assert processed is True
+    assert captured["raw"] == "@codex review security --full\nFocus auth/session handling."

tests/test_jobs.py

@@ -3,6 +3,7 @@ from __future__ import annotations
 from sqlalchemy.exc import IntegrityError
 
 from gitea_codex_bot.db import get_session_factory
+from gitea_codex_bot.models import ReviewJob
 from gitea_codex_bot.services.jobs import cooldown_remaining_seconds, enqueue_job, persist_webhook_event
 from gitea_codex_bot.types import ParsedCommand
 
@@ -19,7 +20,16 @@ def test_enqueue_and_cooldown() -> None:
     session_factory = get_session_factory()
     with session_factory() as session:
         cmd = ParsedCommand(name="review", raw="@codex review")
-        enqueue_job(session, repo="acme/repo", pr_number=42, head_sha="abc", trigger_comment_id=100, requested_by="user", command=cmd)
+        enqueue_job(
+            session,
+            repo="acme/repo",
+            pr_number=42,
+            head_sha="abc",
+            trigger_comment_id=100,
+            trigger_comment_body="@codex review",
+            requested_by="user",
+            command=cmd,
+        )
         remaining = cooldown_remaining_seconds(session, "acme/repo", 42, 60)
     assert remaining >= 0
 
@@ -28,11 +38,48 @@ def test_trigger_comment_unique() -> None:
     session_factory = get_session_factory()
     with session_factory() as session:
         cmd = ParsedCommand(name="review", raw="@codex review")
-        enqueue_job(session, repo="acme/repo", pr_number=7, head_sha="x", trigger_comment_id=321, requested_by="user", command=cmd)
+        enqueue_job(
+            session,
+            repo="acme/repo",
+            pr_number=7,
+            head_sha="x",
+            trigger_comment_id=321,
+            trigger_comment_body="@codex review",
+            requested_by="user",
+            command=cmd,
+        )
         try:
-            enqueue_job(session, repo="acme/repo", pr_number=7, head_sha="x", trigger_comment_id=321, requested_by="user", command=cmd)
+            enqueue_job(
+                session,
+                repo="acme/repo",
+                pr_number=7,
+                head_sha="x",
+                trigger_comment_id=321,
+                trigger_comment_body="@codex review",
+                requested_by="user",
+                command=cmd,
+            )
             duplicate_raised = False
         except IntegrityError:
             duplicate_raised = True
             session.rollback()
     assert duplicate_raised is True
+
+
+def test_enqueue_persists_full_trigger_comment_body() -> None:
+    session_factory = get_session_factory()
+    with session_factory() as session:
+        cmd = ParsedCommand(name="review", raw="@codex review security\nplease focus auth")
+        job = enqueue_job(
+            session,
+            repo="acme/repo",
+            pr_number=55,
+            head_sha="abc123",
+            trigger_comment_id=9191,
+            trigger_comment_body=cmd.raw,
+            requested_by="alice",
+            command=cmd,
+        )
+        stored = session.get(ReviewJob, job.id)
+        assert stored is not None
+        assert stored.trigger_comment_body == "@codex review security\nplease focus auth"

tests/test_review_format.py

@@ -28,3 +28,30 @@ def test_format_result_comment_replaces_existing_marker() -> None:
     assert body.startswith("<!-- codex-review:head_sha=def5678 -->")
     assert "old" not in body.splitlines()[0]
 
+
+def test_format_result_comment_appends_usage_note_for_markdown_comment() -> None:
+    body = format_result_comment(
+        "ff0011",
+        {
+            "markdown_comment": "## Codex Review\n\nLooks fine.",
+            "_meta": {
+                "model": "gpt-5.3-codex",
+                "usage": {"input_tokens": 120, "output_tokens": 45, "total_tokens": 165},
+            },
+        },
+    )
+    assert "_Note: model `gpt-5.3-codex`, input `120`, output `45`, total `165` tokens used._" in body
+
+
+def test_format_result_comment_appends_usage_note_for_fallback_layout() -> None:
+    body = format_result_comment(
+        "ff0011",
+        {
+            "verdict": "correct",
+            "confidence": 0.8,
+            "summary": "No issues.",
+            "findings": [],
+            "_meta": {"model": "gpt-5.3-codex", "usage": {"total_tokens": 88}},
+        },
+    )
+    assert body.endswith("_Note: model `gpt-5.3-codex`, total `88` tokens used._")

tests/test_reviewer_fallback.py

@@ -4,7 +4,7 @@ import httpx
 
 from gitea_codex_bot.config import get_settings
 from gitea_codex_bot.services.repo_config import RepoReviewConfig
-from gitea_codex_bot.services.reviewer import _fallback_review, run_review_for_pr
+from gitea_codex_bot.services.reviewer import _build_prompt, _fallback_review, run_review_for_pr
 from gitea_codex_bot.types import ParsedCommand
 
 
@@ -37,3 +37,21 @@ def test_run_review_for_pr_uses_openai_http_error_in_fallback(monkeypatch) -> No
     assert result["findings"][0]["title"] == "OpenAI review request failed"
     assert "rate_limited" in result["findings"][0]["body"]
     assert any(finding["title"] == "TODO marker in diff" for finding in result["findings"])
+
+
+def test_build_prompt_includes_trigger_message() -> None:
+    pr = type("PR", (), {"html_url": "https://gitea.example/pr/1"})()
+    command = ParsedCommand(name="review", raw="@codex review security\nPlease focus auth.")
+    diff_context = {"truncated": False, "changed_files": ["app.py"], "diff": "diff --git a/app.py b/app.py"}
+    repo_cfg = RepoReviewConfig()
+
+    prompt = _build_prompt(
+        pr,
+        command,
+        diff_context,
+        repo_cfg,
+        changed_file_contents="",
+        test_output=None,
+    )
+
+    assert "Trigger message: @codex review security\nPlease focus auth." in prompt

tests/test_transitions.py

@@ -17,6 +17,7 @@ def test_claim_and_transition() -> None:
             pr_number=314,
             head_sha="deadbeef",
             trigger_comment_id=9901,
+            trigger_comment_body="@codex review",
             requested_by="alice",
             command=ParsedCommand(name="review", raw="@codex review"),
         )
@@ -33,4 +34,4 @@ def test_claim_and_transition() -> None:
     with session_factory() as session:
         loaded = session.execute(select(ReviewJob).where(ReviewJob.id == job.id)).scalar_one()
         assert loaded.status == JobStatus.succeeded
-        assert loaded.result_json is not None
+        assert loaded.result_json is not None

tests/test_webhook.py

@@ -6,8 +6,11 @@ import json
 from typing import Any
 
 from fastapi.testclient import TestClient
+from sqlalchemy import select
 
 from gitea_codex_bot.main import app
+from gitea_codex_bot.db import get_session_factory
+from gitea_codex_bot.models import ReviewJob
 
 
 def _sign(payload: bytes) -> str:
@@ -79,6 +82,10 @@ def test_webhook_accepts_review_and_queues(monkeypatch) -> None:
     assert response.status_code == 200
     assert response.json()["status"] == "queued"
     assert posted_comments
+    session_factory = get_session_factory()
+    with session_factory() as session:
+        queued = session.execute(select(ReviewJob).where(ReviewJob.trigger_comment_id == 111)).scalar_one()
+        assert queued.trigger_comment_body == "@codex review security"
 
 
 def test_webhook_logs_when_no_codex_review_command(monkeypatch) -> None: