* wip(rebuild): core reliability spine
- remove PR-babysitting (ci_autofix + ci_monitor graph + webhook wiring)
- dispatch core: agent/dispatch.py with multitask_strategy=interrupt +
durability=sync + completion webhook; reroute all webhook + plan triggers;
drop the racy in-process lock + is_thread_active busy-check
- completion webhook: agent/completion.py + /webhooks/run-complete loopback
route for failure/timeout replies (idempotent)
Co-authored-by: open-swe[bot]
* feat(rebuild): async tools, reconcile, shared http timeouts, assembly tuning
Parallel batch on top of the reliability spine:
- async-ify all 24 tools (drop asyncio.run; requests->httpx); re-implement the
http_request/fetch_url SSRF + DNS-rebinding defense httpx-natively and harden
the IP check to 'not is_global' (+ IPv4-mapped unwrap)
- reconcile.py: stale pending-run sweep (threads.search -> per-thread runs.list
-> cancel_many), wired into the scheduler graph via task='reconcile'
- shared DEFAULT_HTTP_TIMEOUT (agent/utils/http.py) on every bare
httpx.AsyncClient() across utils/dashboard/webapp/middleware
- run budget: MODEL_CALL_RECURSION_LIMIT 5000->250
- fix stale OpenAI->Anthropic fallback id (claude-opus-4-5 -> 4-8)
- drop redundant custom repair middleware (deepagents auto-adds PatchToolCalls)
- confirm tool-result eviction + summarization auto-wired via backend
- slim system prompt ~8% (full harness-profile rewrite deferred)
Co-authored-by: open-swe[bot]
* feat(rebuild): harness-profile prompt + split webhooks out of webapp
- prompt.py: own the system prompt via a registered harness profile
(OPEN_SWE_SHARED_BASE, kept neutral so the read-only reviewer/analyzer that
share it stay safe), registered across all 4 providers; per-thread values
stay in construct_system_prompt. Assembled main-agent prompt ~6.8k -> ~3.1k
tokens (~55% smaller); de-duped PR/commit/suite/force-push guidance; dropped
ALL-CAPS markers.
- webapp.py 3325 -> 1890 LOC: moved 14 per-source handlers into
agent/webhooks/{linear,slack,github}.py; webapp re-exports them for the
routes + tests; moved handlers reach shared helpers via the webapp namespace
to preserve the test suite's monkeypatch targets.
Full suite: 1168 passing, lint clean.
Co-authored-by: open-swe[bot]
* Restore MODEL_CALL_RECURSION_LIMIT to 5000 for long-running tasks
Reverts the 250 cap from the run-budget change — long-running tasks legitimately
need many model calls. The notify_step_limit_reached safety net still fires if a
run does hit the cap, so runs end with a signal either way.
Co-authored-by: open-swe[bot]
* fix: address PR review (auth, SSRF, interrupted status, redirect headers)
- completion.py: drop `interrupted` from failure statuses — with
multitask_strategy=interrupt a follow-up ends the prior run as interrupted,
which is healthy, not a failure to report. [open-swe]
- /webhooks/run-complete: shared-secret auth — dispatch appends ?token= when
RUN_COMPLETE_WEBHOOK_SECRET is set; route verifies via hmac.compare_digest.
[corridor-security]
- SSRF: extract the URL validator to agent/utils/url_safety.py and apply it
before server-side image fetches in multimodal.fetch_image_block.
[corridor-security]
- http_request: preserve caller headers/extensions across redirect hops instead
of dropping them on the first hop. [open-swe]
Co-authored-by: open-swe[bot]
* chore: remove REBUILD_PLAN.md (planning doc, not needed in the repo)
Co-authored-by: open-swe[bot]
* fix: fail closed on run-complete webhook auth when secret unset
Corridor follow-up: verify_run_complete_token returns False (not True) when
RUN_COMPLETE_WEBHOOK_SECRET is unset, so the public route is never
unauthenticated. Logs a startup warning when the secret is absent, and dispatch
skips registering the webhook when there's no secret (no rejected callbacks).
Co-authored-by: open-swe[bot]
---------
Co-authored-by: open-swe[bot] <open-swe@users.noreply.github.com>
* feat: chat with your PR on the review page
Add a sandbox-less `chat` graph that answers questions about a single PR
from its diff, the published review findings, and read-only GitHub access.
- agent/chat.py: deepagents graph, no sandbox (default StateBackend, file
mutation + execute tools excluded). PR context is seeded as virtual files
under /pr/; a repo-scoped App token is resolved in-graph.
- tools: read_repo_file, search_repo_code, list_review_findings.
- dashboard/review_chat_api.py + routes: per-user chat thread, LangGraph
stream/commands/state/history proxy pinned to the chat assistant, seeds
diff/findings/overview on first run. Gated by repo access.
- UI: Chat tab wired to a chat-scoped StreamProvider (replaces Coming Soon).
* feat: admin setting for review-chat default model
Add a 'Open SWE Review Chat' default to team settings (default_chat_model /
default_chat_reasoning_effort). get_team_default_model("chat") inherits the
Agent default when unset; the chat graph resolves through it. Admin RolePicker
gains an 'Agent default' inherit option that clears the override.
* feat: multi-conversation review chat (tabs, new chat, history)
Replace the single per-PR chat thread with multiple per-user conversations:
- threads minted client-side; first message persists with a title derived
from the prompt.
- list + delete endpoints; chat panel gets a tab strip (history), new-chat
(+), close (x), refresh, an intro greeting, and suggested prompts.
- get_review_chat now returns availability only (ids are client-minted).
* ui fixes
* ui: review-chat history dropdown, full-width AI replies, resizable side panel
* fix(review-chat): enforce per-user thread ownership on proxy endpoints; reseed PR context on head change
---------
Co-authored-by: open-swe[bot] <open-swe@users.noreply.github.com>
* feat: CI auto-fix and PR babysitting for agent PRs
Watch CI failures and review feedback on PRs Open SWE opened, then dispatch
confidence-gated fix runs on the originating agent thread. Adds CI webhook
ingestion (check_run/check_suite/workflow_run/status), a per-PR @open-swe
autofix on|off toggle, auto-response to review comments, and a polling
ci_monitor graph that also flags merge conflicts. Gated by the existing
autofix_mode/trigger_mode settings, the enabled-repos opt-in, base-branch and
human-commit skip rules, dedupe, and a per-PR attempt cap.
Co-authored-by: open-swe[bot] <open-swe@users.noreply.github.com>
* fix: address review feedback on CI auto-fix
- Security: gate the no-mention review-feedback path on author trust —
require a trusted author_association (OWNER/MEMBER/COLLABORATOR) plus a
GitHub write/maintain/admin permission check before dispatching a
write-capable agent run, preventing privilege escalation from
read/triage/outside reviewers.
- Auth: reuse the originating PR thread's source + login/email when
dispatching fix runs so the GitHub-token resolver authenticates them in
non-bot-token deployments (bespoke github_ci source failed to resolve).
- Docs: document the Commit statuses: Read-only permission required for the
Status webhook event.
Co-authored-by: open-swe[bot] <open-swe@users.noreply.github.com>
---------
Co-authored-by: open-swe[bot] <open-swe@users.noreply.github.com>
* feat: out-of-process usage-snapshot builder (Phase 1)
Move all usage-tab compute off the run-serving HTTP process (the #1434 bug
class). Read path is now a pure cache read with a typed computing placeholder
on cold miss; a dedicated usage_snapshot graph + global ~10min cron rebuild
every period's snapshot out-of-process, wrapped in asyncio.timeout and gated by
USAGE_SNAPSHOT_CRON_ENABLED. Lifespan makes one fire-and-forget scheduling call,
never a retained/looping task.
* fix: address PR review on usage-snapshot builder
- admin guard on /admin/usage/rebuild (was any logged-in user)
- cap lifespan loopback calls with asyncio.timeout(5) so a startup hang
can't block boot
- memoize cron registration so steady-state requests skip the loopback
check; reap duplicate crons from concurrent-replica races
- thread the computing flag through the usage payload too
* feat: add scheduled web agents
Co-authored-by: open-swe[bot] <215916821+open-swe[bot]@users.noreply.github.com>
* fix: secure scheduled agent repositories
Co-authored-by: open-swe[bot] <215916821+open-swe[bot]@users.noreply.github.com>
* feat: rebuild scheduled agents as Automations tab
Scrap the inline ScheduledAgentsPanel and replace it with a dedicated
Automations tab: sidebar nav entry, list view with stat cards + empty
state, and a full editor (name, Active toggle, repo, scheduled trigger
picker, agent instructions + model).
* fix: clear collapsed-sidebar button on mobile in Automations
* fix: allow clearing automation repo on update
---------
Co-authored-by: open-swe[bot] <215916821+open-swe[bot]@users.noreply.github.com>
* fix: reset stale sandbox creation sentinel
Co-authored-by: Johannes du Plessis <51395795+johannes117@users.noreply.github.com>
* fix: treat SANDBOX_CREATING as a timestamped cross-process lock
Only reset the sentinel when proven stale (older than the creation
timeout); otherwise wait for the worker that holds the lock so a
concurrent run does not create a duplicate sandbox.
* feat(analyzer): outcomes dataset + bootstrap/continual split via skills
Rename the review_style_analyzer graph to `analyzer` and split it into two
modes, plus capture reviewer finding outcomes for continual learning.
- Outcomes dataset: upsert resolved-by-commit (positive), dismissed (false
positive), and GitHub/Slack thumbs findings into a single LangSmith dataset
(openswe-reviewer-outcomes), keyed deterministically per finding+source.
Emit points wired into update_finding, resolve_finding_thread, and the
GitHub/Slack reaction handlers.
- Two playbooks delivered as deepagents skills (bootstrap-repo-analysis,
continual-learning), served as virtual files via a CompositeBackend /skills/
route + StateBackend (seeded into the run files channel at invoke time, never
written to the sandbox). Mode is set by the launcher; continual runs fall
back to the GitHub App installation token.
- Split launcher into start_bootstrap_analysis + start_continual_run; register
a per-repo nightly continual-learning cron when bootstrap completes.
- New read_finding_outcomes tool feeds confirmed/dismissed findings back to the
continual playbook.
Tests for outcome label mapping, skills helper, and cron idempotency.
* fix(analyzer): anchor continual cron runs to a real thread_id
The nightly continual-learning cron is threadless, and get_analyzer
early-returns an empty agent when configurable.thread_id is missing — so
every cron-launched run no-op'd before reading outcomes or saving a refined
prompt. Include the repo's deterministic analyzer thread_id in the continual
run configurable so the run executes; the threadless run carries no message
history, so nightly runs don't accumulate context.
* refactor(analyzer): move cron lifecycle calls out of the review-styles store
Drop the inline `analyzer_cron` imports from review_styles.py (added only to
dodge a circular import) by relocating the cron-trigger calls to the layer
above the store: registration to the save_review_style tool (after a prompt is
saved) and removal to the dashboard delete route. review_styles.py is now a
pure store again with top-level imports only.
* refactor: hoist reviewer_outcomes imports to module level
Move the two inline emit_finding_status_outcome imports introduced in this PR
(update_finding, resolve_finding_thread) to top-level imports. reviewer_outcomes
only depends on langsmith, so there is no circular import to avoid.
---------
Co-authored-by: open-swe[bot] <open-swe@users.noreply.github.com>
* feat: tune reviewer for precision — web/wiki tools + recalibrated prompt
Reviewer agent now has web_search, fetch_url, and http_request alongside the
finding tools, so it can verify library semantics and consult the DeepWiki
auto-generated wiki for public repos (https://deepwiki.com/<owner>/<repo>)
before flagging cross-file or architectural concerns.
Prompt rewritten to push precision over recall:
- explicit severity ladder pushing reviews toward bimodal high/low instead of
defaulting to medium
- ≤200-char description target (gold set averages ~186 chars; we were at ~436)
- mandatory docs / wiki / code lookup before flagging concurrency, security,
or perf — the three categories that dominated false positives
- "do not flag" list covering compiler/linter-catchable nits, speculative
claims without a concrete attacker/interleaving/scale, style preferences
the codebase doesn't share, and test-quality nits on non-test diffs
- smart file-selection guidance for large PRs (deprioritize generated /
vendored / pure-rename hunks)
Eval config switched to openai:gpt-5.5 + high reasoning effort for the next
benchmark run.
* trim prompt
* subagent prompting
* confidence ratings
* added medium
* enforce confidence threshold
* .
* reviewer: precision-tuned prompt + drop confidence gate
Rewrites the reviewer system prompt around a defensibility bar (anchor +
failure mode + maintainer wouldn't say "not a bug"), an explicit do-not-file
list (style nits, speculation, scope-policing, same-bug fan-out), and a
checklist of 10 bug archetypes drawn from a per-PR audit of the eval golden
set. The audit showed 145 FPs in the last eval split ~28% speculative, ~26%
style-nit, ~31% real-but-unscored (mostly same-archetype fan-out); the new
prompt targets each class directly.
Confidence is still recorded on every finding for post-hoc calibration but
no longer gates publication — the audit showed the gate was a no-op (agent
self-rated 65% of findings "high" regardless), and the prompt's defensibility
bar is the actual discipline. Drops CONFIDENCE_ORDER, CONFIDENCE_THRESHOLD,
the confidence_threshold kwarg on filter_findings_for_publish, the
confidence_filtered score_mode, and the min_confidence kwarg on the eval
target's _extract_comments — all dead once the gate is gone.
Also removes the "informational" severity tier from the Severity enum,
SEVERITY_ORDER, and all validators / tests / docstrings. It was reserved for
FYI observations the dataset never rewards.
* benchmax
* adding google provider
* slight steering
* tuning
* more tuning
* fix
* cleanup
* reducing overfitting
* Add per-repo review style profiles and inject them into the reviewer.
Dashboard users can analyze historical PR review feedback per repository,
edit the resulting style guide, and have it loaded from LangGraph Store at
reviewer runtime (including Martian eval runs) keyed by owner/name.
Co-authored-by: Cursor <cursoragent@cursor.com>
* Fix review style job errors leaking exception details to clients.
Return generic dashboard messages while logging full stack traces server-side.
Co-authored-by: Cursor <cursoragent@cursor.com>
---------
Co-authored-by: open-swe[bot] <open-swe@users.noreply.github.com>
Co-authored-by: Cursor <cursoragent@cursor.com>
* feat: add reviewer graph + eval target wiring
- New `reviewer` graph (`agent/reviewer.py`) registered in langgraph.json
alongside the main `agent` graph. Reuses the same sandbox lifecycle,
GH proxy auth, and middleware primitives from `agent.server`, but with
a narrower tool set, a reviewer-specific system prompt, no
commit/push, and the `task` (subagent) tool stripped via
`_ToolExclusionMiddleware` so review stays in one context.
- New `github_comment` tool: agents call it once per issue with
`(file, line, body, severity)` and the eval scores those calls
against golden comments.
- `ensure_no_empty_msg` middleware (the no_op nudge) is intentionally
*not* on the reviewer's stack — that middleware exists to enforce the
main agent's "always finalize via Slack/Linear/PR" contract, which
the reviewer doesn't have. The main agent's behavior is unchanged.
- `evals/reviewer/target.py`: send PR info as a user message, extract
every `github_comment` tool call (multiple expected per review) into
the run output.
- `evals/reviewer/judge.py`: per-example evaluator now returns a list
of metrics under `{"results": [...]}` so LangSmith averages each
numeric key (f1/precision/recall/tp/fp/fn) across the experiment in
the UI. Dropped the broken `aggregate_pr` summary evaluator that
reached for an attribute that doesn't exist on `RunTree`.
- `evals/reviewer/run_eval.py`: `--limit` now slices the dataset via
`client.list_examples(limit=N)` since `aevaluate` doesn't accept
`max_examples`.
- Makefile: `dev` and `run` targets now use `uv run` so they work
without an activated venv.
* resolve comments
---------
Co-authored-by: open-swe[bot] <open-swe@users.noreply.github.com>
* feat: New multi agent GitHub based workflow
* fix: read plan from gh issue in init issue node
* format
* cr
* fix how new planner threads are started, allow for creating new issues
* fix: read tasks from gh issue inside manager when taking any action
* cr
* update planner
* feat: port intrerrupt node to planner
* add interrupt routing in graph
* update programmer nodes and state
* read messages from issue in programmer
* lint fix
* tag issue when opening PR
* better pr title and body generation prompts
* start runs from planner interrupt and update issue with task plan
* fix lg server build issues
* cr
* fix cloning and pushing
* cr
* cr
* feat: v0 copy (#194)
* feat: v0 copy
* cr
* all pages 'working'
* cr
* cr
* cr
* cr
