researchloop 0.2.0__tar.gz → 0.3.1__tar.gz

{researchloop-0.2.0 → researchloop-0.3.1}/CLAUDE.md

@@ -8,18 +8,19 @@ ResearchLoop is an automated research sprint platform for HPC clusters. It orche
 
 Two processes:
 
-1. **Orchestrator** (`researchloop serve`) — FastAPI server that manages studies/sprints in SQLite, submits jobs via SSH, receives webhooks from runners, stores artifacts. Also serves the web dashboard and handles Slack events.
+1. **Orchestrator** (`researchloop serve`) — FastAPI server that manages studies/sprints in SQLite, submits jobs via SSH, receives webhooks from runners, stores artifacts. Also serves the web dashboard and handles Slack events. Has no Claude CLI dependency.
 2. **Sprint Runner** — runs inside each SLURM/SGE job on HPC. Self-contained bash scripts chain `claude -p` calls through a pipeline (research → red-team → fix → report → summarize), then upload artifacts and send a completion webhook.
 
 Key design decisions:
 
-- All AI work runs on HPC, never on the orchestrator (except Slack conversations and auto-loop idea generation, which use `claude -p` locally with restricted tools)
-- `claude -p --output-format stream-json` for sprint steps (enables live progress), `--output-format json` for conversations
+- All AI work runs on HPC; the orchestrator never invokes `claude`
+- `claude -p --output-format stream-json` for sprint steps (enables live progress)
 - SSH to HPC login nodes for sbatch/squeue/scancel/qsub/qdel
 - Job completion via per-sprint webhook tokens (runner → orchestrator), SSH polling as fallback
 - SQLite (aiosqlite, WAL mode) for metadata, with a `settings` table for persistent config (signing key, password hash)
 - Jinja2 templates for all prompts and job scripts — prompts are pre-rendered by the orchestrator and embedded as base64 in the job script
 - Auto-loop sprints generate their own ideas on the cluster (where Claude is authenticated) rather than on the orchestrator
+- Slack integration is notification + structured slash-style commands only (`sprint run`, `sprint list`, `loop start`, `help`); free-form Q&A was removed
 - Context hierarchy: global → cluster → study (inline text + file paths at each level)
 
 ## Tech stack
@@ -51,11 +52,10 @@ researchloop/
     models.py           — SprintStatus enum, Sprint/Study/AutoLoop dataclasses, generate_sprint_id(), format_sprint_dirname()
     orchestrator.py     — Orchestrator class + create_app() FastAPI factory (API + Slack + dashboard)
     credentials.py      — CLI credential storage (~/.config/researchloop/credentials.json) for remote orchestrator auth
-    auth.py             — check_claude_auth_async() helper for verifying Claude CLI auth status
   db/
     __init__.py
     database.py         — async SQLite wrapper (WAL mode, auto-migrations, fetch_one/fetch_all/execute)
-    migrations.py       — CREATE TABLE statements (7 tables: studies, sprints, auto_loops, artifacts, slack_sessions, events, settings) + indexes + incremental column migrations
+    migrations.py       — CREATE TABLE statements (7 tables: studies, sprints, tweaks, auto_loops, artifacts, events, settings) + indexes + incremental column migrations
     queries.py          — async CRUD functions (all take Database as first arg, return dicts)
   clusters/
     __init__.py
@@ -95,7 +95,6 @@ researchloop/
     base.py             — BaseNotifier ABC (notify_sprint_started/completed/failed)
     ntfy.py             — NtfyNotifier (ntfy.sh push notifications)
     slack.py            — SlackNotifier (chat:write + files:write) + verify_slack_signature()
-    conversation.py     — ConversationManager (Slack threads → Claude sessions via --resume, action execution, markdown→Slack conversion)
     router.py           — NotificationRouter (fan-out to all configured notifiers)
   dashboard/
     __init__.py
@@ -116,7 +115,7 @@ researchloop/
 
 ## Database
 
-SQLite with 8 tables: `studies`, `sprints`, `tweaks`, `auto_loops`, `artifacts`, `slack_sessions`, `events`, `settings`. Schema in `db/migrations.py`. All queries in `db/queries.py` use parameterized SQL and return plain dicts.
+SQLite with 7 tables: `studies`, `sprints`, `tweaks`, `auto_loops`, `artifacts`, `events`, `settings`. Schema in `db/migrations.py`. All queries in `db/queries.py` use parameterized SQL and return plain dicts. (An older `slack_sessions` table is dropped by the migration if present.)
 
 Key columns:
 - `sprints.webhook_token` — per-sprint token for webhook auth (generated at creation)
@@ -148,7 +147,7 @@ Key columns:
 - CSRF protection: HMAC-based tokens derived from session token + signing secret, checked on all mutating dashboard POST routes
 - Dashboard refresh: pulls live status from cluster via SSH (reads logs, progress.md, output.log, report.md, findings.md, summary.txt, idea.txt, checks for PDF)
 - Slack events: deduplication via event_id set, signature verification, background task processing (return 200 immediately), bot message filtering
-- Slack conversation: thread → session mapping in DB, context building with study/sprint info, action execution via [ACTION: ...] tags
+- Slack commands: `sprint run`, `sprint list`, `loop start`, `help` (no free-form chat — orchestrator does not run Claude locally)
 - Auto-loop: sprint idea=None → job script generates idea on cluster → idea.txt read back via SSH/webhook
 - CLI auth: `researchloop connect` gets a bearer token via /api/auth, stored in ~/.config/researchloop/credentials.json with 600 permissions
 - CLI auto-reauth: on 401, prompts for password, gets new token, saves it
@@ -156,7 +155,7 @@ Key columns:
 
 ## Testing
 
-339 unit tests covering: models, config parsing, database operations, all query functions, SLURM scheduler (mock SSH), SGE scheduler (mock SSH), local scheduler (real subprocesses), study/sprint managers, auto-loop controller (with mock claude), notification router, Slack notifier + signature verification + conversation manager + Slack events API, FastAPI API endpoints (TestClient), dashboard routes + auth + setup + CSRF, CLI commands (CliRunner), runner output parsing, and template rendering.
+Unit tests cover: models, config parsing, database operations, all query functions, SLURM scheduler (mock SSH), SGE scheduler (mock SSH), local scheduler (real subprocesses), study/sprint managers, auto-loop controller, notification router, Slack notifier + signature verification + Slack events API, FastAPI API endpoints (TestClient), dashboard routes + auth + setup + CSRF, CLI commands (CliRunner), runner output parsing, and template rendering.
 
 Integration tests (in tests/integration/) use a Docker SLURM container to test real job submission.
 

{researchloop-0.2.0 → researchloop-0.3.1}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: researchloop
-Version: 0.2.0
+Version: 0.3.1
 Summary: Automated research sprint platform for HPC clusters
 License: MIT
 License-File: LICENSE
@@ -38,6 +38,8 @@ Description-Content-Type: text/markdown
 [![Python 3.10+](https://img.shields.io/badge/python-3.10%2B-blue.svg)](https://www.python.org/downloads/)
 [![License: MIT](https://img.shields.io/badge/License-MIT-green.svg)](https://opensource.org/licenses/MIT)
 
+<img width="720" height="456" alt="mmlu-combined" src="https://github.com/user-attachments/assets/6d1d495f-1078-4f81-9f8a-bb1792ea3905" />
+
 ---
 
 ResearchLoop submits AI-powered research experiments to your SLURM or SGE cluster, then reports back the results. You describe a research idea in natural language, it handles the rest: submitting the job, running a multi-step pipeline with [Claude Code](https://docs.anthropic.com/en/docs/claude-code), red-teaming the results, generating a report, and notifying you when it's done.
@@ -120,11 +122,13 @@ Browse to `/dashboard/` to see all your studies, sprints, and loops. Submit new
 
 ### Slack bot
 
-Chat with the bot to start sprints, check status, or discuss research ideas. The bot maintains conversation context across a thread, so you can have a back-and-forth about what to try next.
+Get sprint notifications in your Slack channel and run commands from a thread:
 
 ```
-You: What should I investigate next based on the results from sp-a3f7b2?
-Bot: Based on the findings, I'd suggest... [ACTION: sprint_run {"study": "my-project", "idea": "..."}]
+sprint run my-project "investigate feature X under condition Y"
+sprint list
+loop start my-project 5
+help
 ```
 
 See the [Slack setup guide](https://researchloop.github.io/researchloop/slack/) for configuration.
@@ -187,7 +191,7 @@ Full docs at **[researchloop.github.io/researchloop](https://researchloop.github
 - [Configuration reference](https://researchloop.github.io/researchloop/configuration/) -- all TOML options and environment variables
 - [Deployment guide](https://researchloop.github.io/researchloop/deployment/) -- Docker, Fly.io, SSH key setup
 - [Dashboard guide](https://researchloop.github.io/researchloop/dashboard/) -- web UI features and authentication
-- [Slack integration](https://researchloop.github.io/researchloop/slack/) -- setup, commands, conversational mode
+- [Slack integration](https://researchloop.github.io/researchloop/slack/) -- setup, commands, notifications
 - [CLI reference](https://researchloop.github.io/researchloop/cli/) -- all commands with examples
 - [Security](https://researchloop.github.io/researchloop/security/) -- authentication, CSRF, webhook tokens
 - [Development](https://researchloop.github.io/researchloop/development/) -- contributing, testing, architecture

{researchloop-0.2.0 → researchloop-0.3.1}/README.md

@@ -7,6 +7,8 @@
 [![Python 3.10+](https://img.shields.io/badge/python-3.10%2B-blue.svg)](https://www.python.org/downloads/)
 [![License: MIT](https://img.shields.io/badge/License-MIT-green.svg)](https://opensource.org/licenses/MIT)
 
+<img width="720" height="456" alt="mmlu-combined" src="https://github.com/user-attachments/assets/6d1d495f-1078-4f81-9f8a-bb1792ea3905" />
+
 ---
 
 ResearchLoop submits AI-powered research experiments to your SLURM or SGE cluster, then reports back the results. You describe a research idea in natural language, it handles the rest: submitting the job, running a multi-step pipeline with [Claude Code](https://docs.anthropic.com/en/docs/claude-code), red-teaming the results, generating a report, and notifying you when it's done.
@@ -89,11 +91,13 @@ Browse to `/dashboard/` to see all your studies, sprints, and loops. Submit new
 
 ### Slack bot
 
-Chat with the bot to start sprints, check status, or discuss research ideas. The bot maintains conversation context across a thread, so you can have a back-and-forth about what to try next.
+Get sprint notifications in your Slack channel and run commands from a thread:
 
 ```
-You: What should I investigate next based on the results from sp-a3f7b2?
-Bot: Based on the findings, I'd suggest... [ACTION: sprint_run {"study": "my-project", "idea": "..."}]
+sprint run my-project "investigate feature X under condition Y"
+sprint list
+loop start my-project 5
+help
 ```
 
 See the [Slack setup guide](https://researchloop.github.io/researchloop/slack/) for configuration.
@@ -156,7 +160,7 @@ Full docs at **[researchloop.github.io/researchloop](https://researchloop.github
 - [Configuration reference](https://researchloop.github.io/researchloop/configuration/) -- all TOML options and environment variables
 - [Deployment guide](https://researchloop.github.io/researchloop/deployment/) -- Docker, Fly.io, SSH key setup
 - [Dashboard guide](https://researchloop.github.io/researchloop/dashboard/) -- web UI features and authentication
-- [Slack integration](https://researchloop.github.io/researchloop/slack/) -- setup, commands, conversational mode
+- [Slack integration](https://researchloop.github.io/researchloop/slack/) -- setup, commands, notifications
 - [CLI reference](https://researchloop.github.io/researchloop/cli/) -- all commands with examples
 - [Security](https://researchloop.github.io/researchloop/security/) -- authentication, CSRF, webhook tokens
 - [Development](https://researchloop.github.io/researchloop/development/) -- contributing, testing, architecture

{researchloop-0.2.0 → researchloop-0.3.1}/docs/index.md

@@ -2,6 +2,10 @@
 
 **Automated AI research sprints on HPC clusters.**
 
+<video autoplay muted loop playsinline width="720" style="max-width:100%;height:auto;border-radius:6px">
+  <source src="assets/mmlu-combined.mp4" type="video/mp4">
+</video>
+
 ---
 
 ResearchLoop automates multi-step AI research pipelines on SLURM and SGE clusters. You describe a research idea, and ResearchLoop submits it to your HPC cluster where [Claude Code](https://docs.anthropic.com/en/docs/claude-code) executes a full research pipeline -- coding, red-teaming, fixing, reporting -- inside a single job. Results are reported back via webhooks, Slack, or push notifications, and you can monitor everything from a web dashboard or the CLI.

{researchloop-0.2.0 → researchloop-0.3.1}/pyproject.toml

@@ -1,6 +1,6 @@
 [project]
 name = "researchloop"
-version = "0.2.0"
+version = "0.3.1"
 description = "Automated research sprint platform for HPC clusters"
 readme = "README.md"
 license = {text = "MIT"}

researchloop-0.3.1/researchloop/__init__.py

@@ -0,0 +1 @@
+__version__ = "0.3.1"

{researchloop-0.2.0 → researchloop-0.3.1}/researchloop/clusters/monitor.py

@@ -6,12 +6,15 @@ import asyncio
 import json
 import logging
 from datetime import datetime, timezone
-from typing import Any
+from typing import TYPE_CHECKING, Any
 
 from researchloop.clusters.ssh import SSHManager
 from researchloop.db import queries
 from researchloop.schedulers.base import BaseScheduler
 
+if TYPE_CHECKING:
+    from researchloop.sprints.manager import SprintManager
+
 logger = logging.getLogger(__name__)
 
 # If a job's heartbeat is older than this many seconds AND the job is not
@@ -28,11 +31,17 @@ class JobMonitor:
         db: Any,
         schedulers: dict[str, BaseScheduler],
         config: Any = None,
+        sprint_manager: SprintManager | None = None,
     ) -> None:
         self.ssh_manager = ssh_manager
         self.db = db
         self.schedulers = schedulers
         self.config = config
+        # Optional: when set, terminal-state transitions go through
+        # sprint_manager.mark_sprint_terminal so the parent auto-loop
+        # advances. None falls back to a direct DB update (used by
+        # minimal test fixtures that don't construct a SprintManager).
+        self.sprint_manager = sprint_manager
         self._polling_task: asyncio.Task[None] | None = None
         self._stop_event = asyncio.Event()
 
@@ -143,12 +152,17 @@ class JobMonitor:
             # Persist the updated status if it changed.
             if status in ("completed", "failed"):
                 try:
-                    await queries.update_sprint(
-                        self.db,
-                        sprint_id,
-                        status=status,
-                        completed_at=datetime.now(timezone.utc).isoformat(),
-                    )
+                    if self.sprint_manager is not None:
+                        await self.sprint_manager.mark_sprint_terminal(
+                            sprint_id, status
+                        )
+                    else:
+                        await queries.update_sprint(
+                            self.db,
+                            sprint_id,
+                            status=status,
+                            completed_at=datetime.now(timezone.utc).isoformat(),
+                        )
                 except Exception:
                     logger.exception(
                         "Failed to update DB status for sprint %s", sprint_id

{researchloop-0.2.0 → researchloop-0.3.1}/researchloop/comms/slack.py

@@ -26,12 +26,10 @@ class SlackNotifier(BaseNotifier):
         bot_token: str,
         channel_id: str | None = None,
         dashboard_url: str | None = None,
-        conversation_manager: Any = None,
     ) -> None:
         self.bot_token = bot_token
         self.channel_id = channel_id
         self.dashboard_url = dashboard_url
-        self._cm = conversation_manager
 
     async def _post_message(
         self,
@@ -121,10 +119,7 @@ class SlackNotifier(BaseNotifier):
             f"*Study:* {study_name}\n"
             f"*Idea:* {idea_trunc}"
         )
-        resp = await self._post_message(msg)
-        ts = resp.get("ts", "")
-        if ts and self._cm:
-            await self._cm.store_bot_message(ts, msg)
+        await self._post_message(msg)
 
     async def notify_sprint_completed(
         self,
@@ -140,11 +135,7 @@ class SlackNotifier(BaseNotifier):
             f"*Study:* {study_name}\n"
             f"*Summary:* {summary_trunc}"
         )
-        resp = await self._post_message(msg)
-        # Store the notification for thread context.
-        ts = resp.get("ts", "")
-        if ts and self._cm:
-            await self._cm.store_bot_message(ts, msg)
+        await self._post_message(msg)
         if pdf_path:
             await self._upload_file(
                 pdf_path,
@@ -162,10 +153,7 @@ class SlackNotifier(BaseNotifier):
         msg = (
             f":x: Sprint *{link}* failed\n*Study:* {study_name}\n*Error:* {error[:500]}"
         )
-        resp = await self._post_message(msg)
-        ts = resp.get("ts", "")
-        if ts and self._cm:
-            await self._cm.store_bot_message(ts, msg)
+        await self._post_message(msg)
 
 
 def verify_slack_signature(

{researchloop-0.2.0 → researchloop-0.3.1}/researchloop/core/orchestrator.py

@@ -15,7 +15,6 @@ from fastapi.responses import JSONResponse
 
 from researchloop.clusters.monitor import JobMonitor
 from researchloop.clusters.ssh import SSHManager
-from researchloop.comms.conversation import ConversationManager
 from researchloop.comms.ntfy import NtfyNotifier
 from researchloop.comms.router import NotificationRouter
 from researchloop.comms.slack import (
@@ -51,7 +50,6 @@ class Orchestrator:
         self.auto_loop: AutoLoopController | None = None
         self.notification_router: NotificationRouter | None = None
         self.job_monitor: JobMonitor | None = None
-        self.conversation_manager: ConversationManager | None = None
 
     # ------------------------------------------------------------------
     # Lifecycle
@@ -108,24 +106,15 @@ class Orchestrator:
             notification_router=self.notification_router,
         )
 
-        # 6b. Conversation manager
-        self.conversation_manager = ConversationManager(
-            self.db, sprint_manager=self.sprint_manager
-        )
-
-        # Wire conversation manager to Slack notifier
-        # so notifications store thread context.
-        if self.config.slack and self.config.slack.bot_token:
-            for n in self.notification_router._notifiers:
-                if isinstance(n, SlackNotifier):
-                    n._cm = self.conversation_manager
-
         # 7. Auto-loop controller
         self.auto_loop = AutoLoopController(
             db=self.db,
             sprint_manager=self.sprint_manager,
             config=self.config,
         )
+        # Late-bind the back-reference so SprintManager.mark_sprint_terminal
+        # can advance the parent loop on every terminal transition.
+        self.sprint_manager.auto_loop = self.auto_loop
 
         # 8. Job monitor
         self.job_monitor = JobMonitor(
@@ -133,6 +122,7 @@ class Orchestrator:
             db=self.db,
             schedulers=self.schedulers,
             config=self.config,
+            sprint_manager=self.sprint_manager,
         )
         await self.job_monitor.start_polling()
 
@@ -419,6 +409,10 @@ def create_app(orchestrator: Orchestrator) -> FastAPI:
                 {"ok": True, "sprint_id": sprint_id, "tweak_id": tweak_id}
             )
 
+        # handle_completion fires auto_loop.on_sprint_complete internally
+        # via mark_sprint_terminal — single chokepoint for terminal-state
+        # transitions, so the loop also advances when the JobMonitor or a
+        # dashboard refresh is the one that detects the terminal status.
         await orchestrator.sprint_manager.handle_completion(
             sprint_id=sprint_id,
             status=status,
@@ -427,10 +421,6 @@ def create_app(orchestrator: Orchestrator) -> FastAPI:
             idea=idea,
         )
 
-        # Trigger auto-loop advancement if applicable.
-        if orchestrator.auto_loop is not None:
-            await orchestrator.auto_loop.on_sprint_complete(sprint_id)
-
         logger.info(
             "Webhook: sprint %s completion processed (status=%s)",
             sprint_id,
@@ -790,33 +780,6 @@ def create_app(orchestrator: Orchestrator) -> FastAPI:
             return
         text_lower = text.lower().strip()
 
-        # Handle "auth status" / "login" commands
-        if any(kw in text_lower for kw in ("auth status", "auth check", "login")):
-            if slack_cfg and slack_cfg.bot_token:
-                from researchloop.core.auth import (
-                    check_claude_auth_async,
-                )
-
-                ok, detail = await check_claude_auth_async()
-                notifier = SlackNotifier(
-                    bot_token=slack_cfg.bot_token,
-                    channel_id=channel,
-                )
-                if ok:
-                    msg = (
-                        ":white_check_mark: Claude is"
-                        f" authenticated on this server ({detail})."
-                    )
-                else:
-                    msg = (
-                        ":information_source: Claude is not"
-                        " authenticated on this server"
-                        " (not required — AI runs on the"
-                        " HPC cluster)."
-                    )
-                await notifier._post_message(msg, thread_ts=thread_ts)
-            return
-
         # Handle "help" command.
         if text_lower == "help":
             if slack_cfg and slack_cfg.bot_token:
@@ -831,7 +794,6 @@ def create_app(orchestrator: Orchestrator) -> FastAPI:
                     "• `sprint list` — list recent sprints\n"
                     "• `loop start <study> <count>`"
                     " — start an auto-loop\n"
-                    "• `auth status` — check Claude auth\n"
                     "• `help` — show this message",
                     thread_ts=thread_ts,
                 )
@@ -894,29 +856,16 @@ def create_app(orchestrator: Orchestrator) -> FastAPI:
                     )
                 return
 
-        # Free-form chat — pass to Claude via ConversationManager.
-        cm = orchestrator.conversation_manager
-        if cm is not None and slack_cfg and slack_cfg.bot_token:
+        # Unrecognized message — point user at the help command.
+        if slack_cfg and slack_cfg.bot_token:
             notifier = SlackNotifier(
                 bot_token=slack_cfg.bot_token,
                 channel_id=channel,
             )
-
-            try:
-                response_text = await cm.handle_message(
-                    thread_ts=thread_ts,
-                    user_text=text,
-                    channel=channel,
-                    bot_token=slack_cfg.bot_token if slack_cfg else None,
-                )
-                await notifier._post_message(response_text, thread_ts=thread_ts)
-            except Exception as exc:
-                logger.exception("Chat handler failed: %s", exc)
-                await notifier._post_message(
-                    "Sorry, something went wrong. Try `help` for available commands.",
-                    thread_ts=thread_ts,
-                )
-
+            await notifier._post_message(
+                "Sorry, I didn't understand that. Try `help` for available commands.",
+                thread_ts=thread_ts,
+            )
         return
 
     # -- Dashboard HTML routes -----------------------------------------