adam/open-swe
1
0
Fork 0
mirror of https://github.com/Sea-Haven-Industries/open-swe.git synced 2026-09-30 16:13:15 +00:00
Code Issues Projects Releases Packages Wiki Activity Actions

fix: stop inferring Slack repos from message text (#1306)

Browse source 
Avoid treating regex-parsed Slack text as authoritative routing or allowlist input so repository mentions remain model context and access is bounded by installation permissions.
This commit is contained in:
Johannes du Plessis 2026-05-15 15:36:44 -07:00 • committed by GitHub
parent 6647c3463e
commit f662ad6587
No known key found for this signature in database
GPG key ID: B5690EEEBB952194
4 changed files with 68 additions and 151 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

5
INSTALLATION.md
View file

@ -223,7 +223,7 @@ ALLOWED_GITHUB_ORGS="langchain-ai,anthropics"
ALLOWED_GITHUB_REPOS="some-user/their-repo,another-org/specific-repo"
```
A webhook is accepted if the repo's org is in `ALLOWED_GITHUB_ORGS` **or** the `owner/repo` is in `ALLOWED_GITHUB_REPOS`. If both are empty, all repos are allowed.
A GitHub or Linear webhook is accepted if the resolved repo's org is in `ALLOWED_GITHUB_ORGS` **or** the `owner/repo` is in `ALLOWED_GITHUB_REPOS`. If both are empty, all repos are allowed. Slack mentions are not rejected from regex-inferred repository text; repository access is bounded by the GitHub App installation permissions.
### Linear (optional)
@ -386,7 +386,8 @@ GITHUB_OAUTH_PROVIDER_ID="" # The provider ID from steps 3a / 4b
# Leave empty to allow all orgs.
ALLOWED_GITHUB_ORGS="" # e.g. "my-org,my-other-org"
# Comma-separated list of specific owner/repo pairs the agent is allowed to operate on.
# A repo is allowed if its org is in ALLOWED_GITHUB_ORGS OR its owner/repo is in ALLOWED_GITHUB_REPOS.
# For GitHub/Linear webhooks, a repo is allowed if its org is in ALLOWED_GITHUB_ORGS OR its owner/repo is in ALLOWED_GITHUB_REPOS.
# Slack mentions are not rejected from regex-inferred repository text; repository access is bounded by GitHub App installation permissions.
# Leave both empty to allow all repos.
ALLOWED_GITHUB_REPOS="" # e.g. "some-user/their-repo,another-org/specific-repo"

47
agent/webapp.py
View file

@ -486,7 +486,6 @@ async def _upsert_slack_thread_repo_metadata(
async def get_slack_repo_config(
message: str,
channel_id: str,
thread_ts: str,
slack_user_id: str | None = None,
@ -494,31 +493,29 @@ async def get_slack_repo_config(
"""Resolve repository configuration for Slack-triggered runs.
Priority:
1. Explicit ``owner/repo`` mention in the message body.
2. Repo carried over from the existing Slack thread's metadata.
3. The triggering user's dashboard ``default_repo`` (if they have a
1. Repo carried over from the existing Slack thread's metadata.
2. The triggering user's dashboard ``default_repo`` (if they have a
profile and their Slack email maps to a known GitHub login).
4. ``SLACK_REPO_*`` env defaults.
3. ``SLACK_REPO_*`` env defaults.
"""
default_owner = SLACK_REPO_OWNER.strip() or DEFAULT_REPO_OWNER
default_name = SLACK_REPO_NAME.strip() or DEFAULT_REPO_NAME
thread_id = generate_thread_id_from_slack_thread(channel_id, thread_ts)
langgraph_client = get_client(url=LANGGRAPH_URL)
repo_config = extract_repo_from_text(message, default_owner=default_owner)
repo_config: dict[str, str] | None = None
if not repo_config:
try:
thread = await langgraph_client.threads.get(thread_id)
thread_repo_config = _extract_repo_config_from_thread(thread)
if thread_repo_config:
repo_config = thread_repo_config
except Exception as exc: # noqa: BLE001
if not _is_not_found_error(exc):
logger.exception(
"Failed to fetch Slack thread %s for repo resolution",
thread_id,
)
try:
thread = await langgraph_client.threads.get(thread_id)
thread_repo_config = _extract_repo_config_from_thread(thread)
if thread_repo_config:
repo_config = thread_repo_config
except Exception as exc: # noqa: BLE001
if not _is_not_found_error(exc):
logger.exception(
"Failed to fetch Slack thread %s for repo resolution",
thread_id,
)
if not repo_config and slack_user_id:
try:
@ -947,7 +944,9 @@ async def process_slack_mention(event_data: dict[str, Any], repo_config: dict[st
prompt = (
"You were mentioned in Slack.\n\n"
f"## Repository\n{repo_config.get('owner')}/{repo_config.get('name')}\n\n"
"## Default Repository Hint\n"
f"{repo_config.get('owner')}/{repo_config.get('name')}\n"
"Use this only if the Slack conversation does not identify a different repository.\n\n"
f"## Triggered by\n{trigger_user}\n\n"
f"## Slack Thread\n- Channel: {channel_id}\n- Thread TS: {thread_ts}\n"
f"- Context starts at: {context_source}\n\n"
@ -1370,15 +1369,7 @@ async def slack_webhook(request: Request, background_tasks: BackgroundTasks) ->
"text": text,
"bot_user_id": bot_user_id,
}
repo_config = await get_slack_repo_config(text, channel_id, thread_ts, slack_user_id=user_id)
if not _is_repo_allowed(repo_config):
logger.warning(
"Rejecting Slack webhook: repo '%s/%s' not in allowlist",
repo_config.get("owner"),
repo_config.get("name"),
)
return {"status": "ignored", "reason": "Repository not in allowlist"}
repo_config = await get_slack_repo_config(channel_id, thread_ts, slack_user_id=user_id)
background_tasks.add_task(process_slack_mention, event_data, repo_config)

14
tests/test_github_issue_webhook.py
View file

@ -485,10 +485,9 @@ def test_slack_webhook_non_pr_review_request_starts_agent(monkeypatch) -> None:
captured: dict[str, object] = {}
async def fake_get_slack_repo_config(
text: str, channel_id: str, thread_ts: str, slack_user_id: str | None = None
channel_id: str, thread_ts: str, slack_user_id: str | None = None
) -> dict[str, str]:
captured["repo_config_request"] = {
"text": text,
"channel_id": channel_id,
"thread_ts": thread_ts,
"slack_user_id": slack_user_id,
@ -507,6 +506,13 @@ def test_slack_webhook_non_pr_review_request_starts_agent(monkeypatch) -> None:
monkeypatch.setattr(slack_utils.time, "time", lambda: 1700000000)
monkeypatch.setattr(webapp, "get_slack_repo_config", fake_get_slack_repo_config)
monkeypatch.setattr(webapp, "process_slack_mention", fake_process_slack_mention)
monkeypatch.setattr(
webapp,
"_is_repo_allowed",
lambda repo_config: (_ for _ in ()).throw(
AssertionError("Slack webhook should not gate inferred repos with allowlists")
),
)
client = TestClient(webapp.app)
response = _post_slack_webhook(
@ -535,10 +541,9 @@ def test_slack_webhook_threaded_followup_uses_parent_thread_ts(monkeypatch) -> N
captured: dict[str, object] = {}
async def fake_get_slack_repo_config(
text: str, channel_id: str, thread_ts: str, slack_user_id: str | None = None
channel_id: str, thread_ts: str, slack_user_id: str | None = None
) -> dict[str, str]:
captured["repo_config_request"] = {
"text": text,
"channel_id": channel_id,
"thread_ts": thread_ts,
"slack_user_id": slack_user_id,
@ -577,7 +582,6 @@ def test_slack_webhook_threaded_followup_uses_parent_thread_ts(monkeypatch) -> N
assert response.status_code == 200
assert response.json()["message"] == "Slack mention queued"
assert captured["repo_config_request"] == {
"text": "<@UBOT> continue on the branch",
"channel_id": "C123",
"thread_ts": "1700000000.000100",
"slack_user_id": "U123",

153
tests/test_slack_context.py
View file

@ -301,7 +301,7 @@ def test_select_slack_context_messages_detects_username_mention() -> None:
assert [item["ts"] for item in selected] == ["1.0", "2.0", "3.0"]
def test_get_slack_repo_config_message_repo_overrides_existing_thread_repo(
def test_get_slack_repo_config_uses_existing_thread_repo(
monkeypatch: pytest.MonkeyPatch,
) -> None:
threads_client = _FakeThreadsClient(
@ -320,27 +320,27 @@ def test_get_slack_repo_config_message_repo_overrides_existing_thread_repo(
webapp, "post_slack_thread_reply", fake_post_slack_thread_reply, raising=False
)
repo = asyncio.run(
webapp.get_slack_repo_config("please use repo:new-owner/new-repo", "C123", "1.234")
)
repo = asyncio.run(webapp.get_slack_repo_config("C123", "1.234"))
assert repo == {"owner": "new-owner", "name": "new-repo"}
assert threads_client.requested_thread_id is None
assert repo == {"owner": "saved-owner", "name": "saved-repo"}
assert threads_client.requested_thread_id == generate_thread_id_from_slack_thread(
"C123", "1.234"
)
assert not posted
def test_get_slack_repo_config_parses_message_for_new_thread(
def test_get_slack_repo_config_new_thread_uses_default(
monkeypatch: pytest.MonkeyPatch,
) -> None:
threads_client = _FakeThreadsClient(raise_not_found=True)
monkeypatch.setattr(webapp, "SLACK_REPO_OWNER", "default-owner")
monkeypatch.setattr(webapp, "SLACK_REPO_NAME", "default-repo")
monkeypatch.setattr(webapp, "get_client", lambda url: _FakeClient(threads_client))
repo = asyncio.run(
webapp.get_slack_repo_config("please use repo:new-owner/new-repo", "C123", "1.234")
)
repo = asyncio.run(webapp.get_slack_repo_config("C123", "1.234"))
assert repo == {"owner": "new-owner", "name": "new-repo"}
assert repo == {"owner": "default-owner", "name": "default-repo"}
def test_get_slack_repo_config_existing_thread_without_repo_uses_default(
@ -352,7 +352,7 @@ def test_get_slack_repo_config_existing_thread_without_repo_uses_default(
monkeypatch.setattr(webapp, "get_client", lambda url: _FakeClient(threads_client))
repo = asyncio.run(webapp.get_slack_repo_config("please help", "C123", "1.234"))
repo = asyncio.run(webapp.get_slack_repo_config("C123", "1.234"))
assert repo == {"owner": "default-owner", "name": "default-repo"}
assert threads_client.requested_thread_id == generate_thread_id_from_slack_thread(
@ -360,127 +360,43 @@ def test_get_slack_repo_config_existing_thread_without_repo_uses_default(
)
def test_get_slack_repo_config_space_syntax_detected(
def test_get_slack_repo_config_ignores_repo_syntax_in_message(
monkeypatch: pytest.MonkeyPatch,
) -> None:
"""repo owner/name (space instead of colon) should be detected correctly."""
threads_client = _FakeThreadsClient(raise_not_found=True)
monkeypatch.setattr(webapp, "get_client", lambda url: _FakeClient(threads_client))
repo = asyncio.run(
webapp.get_slack_repo_config(
"please fix the bug in repo langchain-ai/langchainjs", "C123", "1.234"
)
)
assert repo == {"owner": "langchain-ai", "name": "langchainjs"}
def test_get_slack_repo_config_github_url_extracted(
monkeypatch: pytest.MonkeyPatch,
) -> None:
"""GitHub URL in message should be used to detect the repo."""
threads_client = _FakeThreadsClient(raise_not_found=True)
monkeypatch.setattr(webapp, "get_client", lambda url: _FakeClient(threads_client))
repo = asyncio.run(
webapp.get_slack_repo_config(
"I found a bug in https://github.com/langchain-ai/langgraph-api please fix it",
"C123",
"1.234",
)
)
assert repo == {"owner": "langchain-ai", "name": "langgraph-api"}
def test_get_slack_repo_config_explicit_repo_beats_github_url(
monkeypatch: pytest.MonkeyPatch,
) -> None:
"""Explicit repo: syntax takes priority over a GitHub URL also present in the message."""
threads_client = _FakeThreadsClient(raise_not_found=True)
monkeypatch.setattr(webapp, "get_client", lambda url: _FakeClient(threads_client))
repo = asyncio.run(
webapp.get_slack_repo_config(
"see https://github.com/langchain-ai/langgraph-api but use repo:my-org/my-repo",
"C123",
"1.234",
)
)
assert repo == {"owner": "my-org", "name": "my-repo"}
def test_get_slack_repo_config_explicit_space_syntax_beats_thread_metadata(
monkeypatch: pytest.MonkeyPatch,
) -> None:
"""Explicit repo owner/name (space syntax) takes priority over saved thread metadata."""
threads_client = _FakeThreadsClient(
thread={"metadata": {"repo": {"owner": "saved-owner", "name": "saved-repo"}}}
)
monkeypatch.setattr(webapp, "get_client", lambda url: _FakeClient(threads_client))
repo = asyncio.run(
webapp.get_slack_repo_config(
"actually use repo langchain-ai/langchainjs today", "C123", "1.234"
)
)
repo = asyncio.run(webapp.get_slack_repo_config("C123", "1.234"))
assert repo == {"owner": "langchain-ai", "name": "langchainjs"}
assert repo == {"owner": "saved-owner", "name": "saved-repo"}
def test_get_slack_repo_config_github_url_beats_thread_metadata(
def test_get_slack_repo_config_applies_profile_default_repo(
monkeypatch: pytest.MonkeyPatch,
) -> None:
"""A GitHub URL in the message takes priority over saved thread metadata."""
threads_client = _FakeThreadsClient(
thread={"metadata": {"repo": {"owner": "saved-owner", "name": "saved-repo"}}}
)
threads_client = _FakeThreadsClient(thread={"metadata": {}})
async def fake_get_slack_user_info(user_id: str) -> dict:
return {"profile": {"email": "mason@example.com"}}
def fake_resolve_login_from_email(email: str | None) -> str | None:
return "mason"
async def fake_get_profile_default_repo(login: str | None) -> dict[str, str] | None:
assert login == "mason"
return {"owner": "profile-owner", "name": "profile-repo"}
monkeypatch.setattr(webapp, "get_client", lambda url: _FakeClient(threads_client))
monkeypatch.setattr(webapp, "get_slack_user_info", fake_get_slack_user_info)
monkeypatch.setattr(webapp, "resolve_login_from_email", fake_resolve_login_from_email)
monkeypatch.setattr(webapp, "get_profile_default_repo", fake_get_profile_default_repo)
repo = asyncio.run(
webapp.get_slack_repo_config(
"I found a bug in https://github.com/langchain-ai/langgraph-api",
"C123",
"1.234",
)
)
repo = asyncio.run(webapp.get_slack_repo_config("C123", "1.234", slack_user_id="U123"))
assert repo == {"owner": "langchain-ai", "name": "langgraph-api"}
def test_get_slack_repo_config_repo_name_only_defaults_org(
monkeypatch: pytest.MonkeyPatch,
) -> None:
"""repo:name without org should default owner to langchain-ai."""
threads_client = _FakeThreadsClient(raise_not_found=True)
monkeypatch.setattr(webapp, "get_client", lambda url: _FakeClient(threads_client))
repo = asyncio.run(
webapp.get_slack_repo_config("fix bug in repo:langchainplus", "C123", "1.234")
)
assert repo == {"owner": "langchain-ai", "name": "langchainplus"}
def test_get_slack_repo_config_repo_name_only_space_syntax(
monkeypatch: pytest.MonkeyPatch,
) -> None:
"""repo name (space syntax, no org) should default owner to langchain-ai."""
threads_client = _FakeThreadsClient(raise_not_found=True)
monkeypatch.setattr(webapp, "get_client", lambda url: _FakeClient(threads_client))
repo = asyncio.run(webapp.get_slack_repo_config("fix bug in repo open-swe", "C123", "1.234"))
assert repo == {"owner": "langchain-ai", "name": "open-swe"}
assert repo == {"owner": "profile-owner", "name": "profile-repo"}
def _setup_slack_mention_fakes(
@ -609,6 +525,11 @@ def test_process_slack_mention_creates_thread_first_run_with_trace_reply(
assert "multitask_strategy" not in kwargs
assert kwargs["config"]["configurable"]["slack_thread"]["thread_ts"] == thread_ts
prompt_block = kwargs["input"]["messages"][0]["content"][0]
assert "## Default Repository Hint\nlangchain-ai/open-swe" in prompt_block["text"]
assert (
"Use this only if the Slack conversation does not identify a different repository."
in (prompt_block["text"])
)
assert prompt_block["text"].count("## Slack Thread") == 1
assert f"Thread TS: {thread_ts}" in prompt_block["text"]
assert "## Latest Mention Request\ncontinue on the branch" in prompt_block["text"]