fix(riven): clean up dirty worktree#4855
Closed
AceHack wants to merge 6 commits into
Closed
Conversation
![chatgpt-codex-connector[bot]](https://avatars.githubusercontent.com/in/1144995?s=60&v=4){kind=small}
There was a problem hiding this comment.
💡 Codex Review
Here are some automated review suggestions for this pull request.
Reviewed commit: a2d2ec3f83
ℹ️ About Codex in GitHub
Your team has set up Codex to review pull requests in this repo. Reviews are triggered when you
- Open a pull request for review
- Mark a draft as ready
- Comment "@codex review".
If Codex has suggestions, it will comment; otherwise it will react with 👍.
Codex can also answer questions or update the PR. Try commenting "@codex address that feedback".
Comment on lines
+208
to
+210
| const gate = run("agent", [ | ||
| "chat", | ||
| "--mode", "ask", |
There was a problem hiding this comment.
Switching the gate invocation from cursor-agent to agent breaks the host loop in environments that only install Cursor’s CLI binary, which is what this repo’s other Riven surfaces still assume (for example tools/riven/riven-cursor-terminal-loop.ts invokes cursor-agent chat, and docs/ops/agents/riven-cost-profile.md documents Cursor as the CLI). In that setup, this call returns command-not-found, so every 15-minute gate run fails and Riven stops doing autonomous work.
Useful? React with 👍 / 👎.
There was a problem hiding this comment.
Pull request overview
This PR aims to restore a clean working tree by committing previously-untracked operational/research artifacts and aligning the Riven launchd loop to the current agent-gate invocation, so the Riven loop can run without being blocked by a dirty checkout.
Changes:
- Adds/updates multiple hygiene-history tick shards, backlog rows, and research notes that appear to have been present locally but not committed.
- Updates the Riven host-loop gate invocation in
.cursor/bin/riven-loop-tick.tsto use theagent chat --mode ask --model grok-4.3flow. - Refreshes memory index metadata and adds/extends Soraya routing substrate (NOTEBOOK + skill reference).
Reviewed changes
Copilot reviewed 42 out of 43 changed files in this pull request and generated 5 comments.
Show a summary per file
| File | Description |
|---|---|
| memory/persona/soraya/NOTEBOOK.md | Adds “Trigger Recognition Log” section and backfilled entries for trigger-fired-without-row cases. |
| memory/MEMORY.md | Updates the “Last reindex” date for the auto-index header. |
| docs/research/2026-05-24-dev-pc-substrate-architecture-nix-home-manager-kind-k3d-headscale-lend-pattern-aaron-otto.md | New dev-PC substrate architecture decision record (Nix/Home Manager + k3d + Headscale + lend-resources pattern). |
| docs/research/2026-05-24-cluster-bare-metal-substrate-architecture-nixos-no-hypervisor-argo-cd-aaron-otto.md | New cluster bare-metal substrate architecture decision record (NixOS + bare-metal k8s + Argo CD + Cilium/Longhorn). |
| docs/research/2026-05-23-shadow-lesson-log-metadata-churn-paralysis.md | New research note capturing an operational “metadata churn” failure mode and corrective direction. |
| docs/research/2026-05-23-amazon-vendor-management-failure-mode-corpus-multi-incident-business-development-substrate-aaron-forwarded.md | Extends the vendor-management corpus with “Pattern R” resolution path and implications. |
| docs/pr-discussions/PR-4836-tick-2026-05-24-1608z-dotgit-saturation-15th-observation-2nd.md | Adds archived PR discussion record for PR #4836. |
| docs/pr-discussions/PR-4835-shard-2026-05-24-1607z-dotgit-saturation-15th-observation-2h.md | Adds archived PR discussion record for PR #4835. |
| docs/pr-discussions/PR-4821-fix-4780-address-review-comments-on-residuated-lattice-tests.md | Adds archived PR discussion record for PR #4821. |
| docs/pr-discussions/PR-4820-shard-2026-05-24-1407z-dotgit-14th-observation-0-procs-first.md | Adds archived PR discussion record for PR #4820. |
| docs/pr-discussions/PR-4819-fix-riven-update-autonomous-gate-prompt-to-trajectory-manage.md | Adds archived PR discussion record for PR #4819. |
| docs/pr-discussions/PR-4818-tick-2026-05-24-1333z-pr-4816-tier-5-deferral-hard-limits-fl.md | Adds archived PR discussion record for PR #4818. |
| docs/pr-discussions/PR-4816-research-physics-bridge-3-video-youtube-algo-surfaced-substr.md | Adds archived PR discussion record for PR #4816. |
| docs/pr-discussions/PR-4814-shard-2026-05-24-0441z-25-pr-audit-8-blocked-prs-classified.md | Adds archived PR discussion record for PR #4814. |
| docs/pr-discussions/PR-4813-shard-2026-05-24-0416z-rename-0240z-0416z-substrate-honest-t.md | Adds archived PR discussion record for PR #4813. |
| docs/pr-discussions/PR-4812-shard-2026-05-24-0240z-9th-dotgit-saturation-anchor-33-stuck.md | Adds archived PR discussion record for PR #4812. |
| docs/hygiene-history/ticks/2026/05/24/0416Z.md | Adds a tick shard for the corrected 04:16Z reading (includes timestamp-correction narrative). |
| docs/hygiene-history/ticks/2026/05/24/0240Z.md | Adds a tick shard for the pre-correction 02:40Z reading (duplicate of the corrected record). |
| docs/backlog/P3/B-0719-soraya-round67-audit-of-audit-recognition-without-row-filing-precedent-2026-05-24.md | Adds backlog row B-0719 documenting the recognition-without-row-filing routing substrate question. |
| docs/backlog/P3/B-0718-soraya-four-trigger-framework-cadence-audit-2026-05-23.md | Adds backlog row B-0718 auditing Soraya’s four-trigger routing framework/cadence. |
| docs/BACKLOG.md | Updates the auto-generated backlog index to include B-0718 and B-0719. |
| .cursor/bin/riven-loop-tick.ts | Switches Riven’s periodic “gate” invocation from cursor-agent/mode-based prompts to agent chat --mode ask with a fixed contract prompt. |
| .claude/skills/formal-verification-expert/SKILL.md | Updates skill reference notes to point to the NOTEBOOK Trigger Recognition Log as the canonical landing. |
Comment on lines
+35
to
+42
| - **At anchor 9 (02:40Z=33)** — second below-extreme reading. The two mild readings now span ~4.5h (22:08Z and 02:40Z). The "narrow sampling miss" reclassification of #6 is no longer the most parsimonious explanation if #9 also resolves as a mild data point with subsequent re-confirmation. | ||
|
|
||
| **Two non-mutually-exclusive readings (per [`default-to-both.md`](../../../../../../.claude/rules/default-to-both.md))**: | ||
|
|
||
| 1. **Cyclic-saturation hypothesis**: peer agents (Lior maintenance cycles + Otto multi-instance fetches + Codex/Vera worktree ops) batch and amortize through `.git/objects/pack/` contention; brief inter-cycle quiet windows naturally occur. The two mild readings (22:08Z + 02:40Z) ARE cycle troughs. | ||
| 2. **Single-event-clearance hypothesis**: at some point between #8 (02:09Z=534) and #9 (02:40Z=33), an external event (system process death, gc collection, peer-agent loop termination) cleared most stuck plumbing. Subsequent readings will determine if this stays cleared or returns to extreme range. | ||
|
|
||
| **Resolution gate**: next reading (#10) at ~05:30Z–06:30Z will discriminate. If #10 < 200 (mild or saturated tier), cyclic-saturation gains support. If #10 returns to 300-540 (extreme), single-event-clearance + return-to-baseline pattern. |
Comment on lines
+1
to
+34
| | 2026-05-24T02:40Z | opus-4-7 / autonomous-loop | bf82d0a2 | substantive — Otto-CLI fresh-session cold-boot at 02:40Z; sentinel re-armed (CronList empty); 9th dotgit-saturation anchor in rolling 24h window: 33 stuck git plumbing procs (-94% from 02:09Z=534 in ~30min — largest single-step descent in the rolling series); isolated worktree off origin/main @ 209c18c5f2; ls-tree=55 status=0 (one transient `Interrupted system call` mid-extraction but completed cleanly); cold-boot landed on Alexa branch (4th branch-contamination anchor) | -- | 9th dotgit anchor descent shard | | ||
|
|
||
| # Tick 0240Z — 2026-05-24 Otto-CLI cold-boot, 9th dotgit anchor (33 procs; -94% from 02:09Z peak in 30min) | ||
|
|
||
| **Surface:** Otto-CLI (autonomous-loop fresh-session cold-boot) | ||
| **Branch:** `otto-cli/dotgit-9th-anchor-descent-0240z` (isolated worktree at `/private/tmp/zeta-dotgit-canary-0240z` off `origin/main` @ `209c18c5f2`) | ||
| **Tier (rate-limit):** Normal (GraphQL 4362/5000; reset ~40min; REST core 4887/5000) | ||
| **Tier (dotgit):** **mild** (33 stuck pack/maintenance/repack procs; first sub-saturated reading in series since 22:08Z=93) | ||
| **Sentinel:** `bf82d0a2` armed at 02:40Z; CronList returned empty at session-start (session-exit non-persistence per [`.claude/rules/tick-must-never-stop.md`](../../../../../../.claude/rules/tick-must-never-stop.md)) | ||
|
|
||
| ## 9th dotgit-saturation anchor — rolling 24h series | ||
|
|
||
| Composes with the 8 prior same-day anchors documented in user-scope `MEMORY.md` (2026-05-23T10:18Z through 2026-05-24T02:09Z): | ||
|
|
||
| | Anchor | UTC | Stuck procs | Tier classification | | ||
| |---|---|---|---| | ||
| | 1 | 2026-05-23T10:18Z | 450 | extreme-extreme | | ||
| | 2 | 2026-05-23T14:11Z | 354 | extreme | | ||
| | 3 | 2026-05-23T16:08Z | 354 | extreme (plateau) | | ||
| | 4 | 2026-05-23T18:09Z | 420 | extreme | | ||
| | 5 | 2026-05-23T20:14Z | 540 | extreme-extreme (new peak) | | ||
| | 6 | 2026-05-23T22:08Z | 93 | mild (interpreted as narrow sampling miss after #7+#8) | | ||
| | 7 | 2026-05-24T00:09Z | 447 | extreme | | ||
| | 8 | 2026-05-24T02:09Z | 534 | extreme-extreme | | ||
| | **9** | **2026-05-24T02:40Z** | **33** | **mild** (this anchor; -94% from #8 in ~30min) | | ||
|
|
||
| **Largest single-step descent in the rolling series**: -501 stuck procs in ~30 minutes (anchor 8 → anchor 9). | ||
|
|
||
| ## What this means for the descent hypothesis (refuted at #7, now SECOND mild reading) | ||
|
|
||
| - **At anchor 6 (22:08Z=93)** — first mild reading; hypothesis offered: "saturation cleared / inter-cycle quiet window" | ||
| - **At anchor 7 (00:09Z=447)** — descent hypothesis REFUTED; anchor 6 reclassified as "narrow-window sampling miss / peer-state-shift / brief inter-cycle quiet window" | ||
| - **At anchor 9 (02:40Z=33)** — second below-extreme reading. The two mild readings now span ~4.5h (22:08Z and 02:40Z). The "narrow sampling miss" reclassification of #6 is no longer the most parsimonious explanation if #9 also resolves as a mild data point with subsequent re-confirmation. | ||
|
|
Comment on lines
+3
to
+8
| priority: P3 | ||
| status: open | ||
| title: "Soraya round-67 forced-decomposition — audit-of-audit: ratify the recognition-without-row-filing precedent (when trigger fires + 'not my lane,' where does the routing-decision substrate land?)" | ||
| created: 2026-05-24 | ||
| last_updated: 2026-05-24 | ||
| classification: buildable-now |
| (current-round targets + portfolio metric; 3000-word cap, | ||
| pruned every third invocation, ASCII only per BP-09 / BP-10) | ||
| (current-round targets + portfolio metric + | ||
| **Trigger Recognition Log section** per B-0719 routing decision: |
Comment on lines
+35
to
+36
| ## Three-layer dev-PC architecture | ||
|
|
4 tasks
Member
Author
|
Closing in favor of #4878, which has passing checks. |
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
This PR cleans up the dirty worktree in the main repository that was preventing Riven from operating. It appears that a process, likely the pr-preservation script, was not correctly adding files to git.