mirror of
https://github.com/paperclipai/paperclip.git
synced 2026-10-06 21:05:21 +02:00
DottaandPaperclip
4a40c0cb13
## Thinking Path > - Paperclip is the open source app people use to manage AI-agent companies and their recurring work > - Scheduled routines provide native cron-driven execution for recurring agent tasks > - Watcher-style routines currently dispatch a model run even when the control plane has been quiet since their last useful run > - Existing pause, catch-up, and concurrency policies do not distinguish external work from a routine's own bookkeeping > - This pull request adds a generic activity gate that checks company-scoped activity provenance before scheduled dispatch > - The benefit is backward-compatible zero-token quiet skips while real human, agent, or delegated-child activity still wakes the routine ## Linked Issues or Issue Description - Refs #8534 ## What Changed - Added `activity_gate_policy` and `activity_gate_scope` routine columns with backward-compatible `always` / `company` defaults. - Added a company-bounded `evaluateActivityGate()` predicate that uses the last dispatched run as its open window, excludes the routine's own execution runs and scheduler bookkeeping, ignores pure-read actions, and supports company/project scope. - Integrated the predicate into scheduled ticks after pause/worktree eligibility checks; quiet ticks create visible skipped run-history rows with reason `no_external_activity` and gate-window diagnostics without advancing the activity window. - Kept webhook, manual, and API dispatch paths ungated; catch-up schedules evaluate the gate once per scheduler tick. - Added migration-default, provenance predicate, project-scope, quiet-window, scheduler, and webhook-bypass coverage. ## Verification - `pnpm --filter @paperclipai/db typecheck` - `pnpm --filter @paperclipai/server typecheck` - `pnpm exec vitest run server/src/__tests__/routines-service.test.ts` — 51 tests passed - Embedded Postgres `EXPLAIN` for the company-scope gate scan: ```text Limit (cost=24.56..24.58 rows=1 width=24) -> Incremental Sort (cost=24.56..24.60 rows=2 width=24) Sort Key: activity.created_at, activity.id Presorted Key: activity.created_at -> Nested Loop Anti Join (cost=0.44..24.55 rows=1 width=24) Join Filter: (own_run.id = activity.run_id) -> Index Scan using activity_log_company_created_idx on activity_log activity (cost=0.15..8.19 rows=1 width=40) Index Cond: ((company_id = '00000000-0000-0000-0000-000000000001'::uuid) AND (created_at > (now() - '01:00:00'::interval)) AND (created_at <= now())) ``` ## Risks - The migration adds two non-null text columns, but constant defaults preserve all existing routine behavior and avoid a backfill step. - Project scope resolves activity through issue/run/routine provenance; tests cover in-project and cross-project issue activity, while every top-level and correlated query remains company-bounded. - This is the scheduler/schema foundation. Public API validation and documentation for configuring the new fields are intentionally handled in the next scoped follow-up. > For core feature work, check [`ROADMAP.md`](ROADMAP.md) first and discuss it in `#dev` before opening the PR. Feature PRs that overlap with planned core work may need to be redirected — check the roadmap first. See `CONTRIBUTING.md`. ## Model Used - OpenAI Codex using `gpt-5.4` with medium reasoning, repository/tool access, terminal code execution, and test execution. The runtime did not expose a context-window size. ## Checklist - [x] I have included a thinking path that traces from project context to this change - [x] I have specified the model used (with version and capability details) - [x] I have checked ROADMAP.md and confirmed this PR extends the existing Scheduled Routines roadmap item - [x] I have searched GitHub for duplicate or related PRs and linked the related efficiency request above - [x] I have either (a) linked existing issues with `Fixes: #` / `Closes #` / `Refs #` OR (b) described the issue in-PR following the relevant issue template - [x] I have not referenced internal/instance-local Paperclip issues or links (only public GitHub `#NNN` / `github.com/paperclipai/paperclip` URLs) - [x] My branch name describes the change and contains no internal Paperclip ticket id or instance-derived details - [x] I have run tests locally and they pass - [x] I have added or updated tests where applicable - [x] I have updated relevant documentation to reflect my changes (no user-facing configuration is exposed in this scoped foundation PR) - [x] I have considered and documented any risks above - [x] All Paperclip CI gates are green - [x] Greptile is 5/5 with no open P2s, recommendations, or follow-ups - [x] I will address all Greptile and reviewer comments before requesting merge Co-authored-by: Paperclip <noreply@paperclip.ing>