@yemi33/minions 0.1.1913 → 0.1.1914

package/README.md

@@ -565,14 +565,12 @@ Engine behavior is controlled via `config.json`. Key settings:
 | `autoDecompose` | true | Auto-decompose `implement:large` items into sub-tasks before dispatch |
 | `autoApprovePlans` | false | Auto-approve PRDs without waiting for human approval |
 | `evalLoop` | true | Auto-dispatch review → fix cycles after implementation completes |
-| `evalMaxIterations` | 3 | Max minion review → fix cycles before pausing minion review automation; human feedback, build fixes, and conflict fixes continue |
 | `evalMaxCost` | null | USD ceiling per work item across all eval iterations (null = no limit) |
 | `meetingRoundTimeout` | 600000 (10min) | Timeout per meeting round before auto-advance |
 | `ccModel` | `sonnet` | Model for Command Center and doc-chat (sonnet/haiku/opus) |
 | `ccEffort` | null | Effort level for CC/doc-chat (null/low/medium/high) |
 | `agentEffort` | null | Override effort level for all agent dispatches |
 | `maxTurnsByType` | `{}` | Per-type max-turns override (e.g., `{"explore": 40, "fix": 100}`) |
-| `maxBuildFixAttempts` | 3 | Max auto-fix dispatches per PR before escalating |
 
 Per-type max-turns defaults (when `maxTurnsByType` not set): explore=30, ask=20, review=30, plan=30, decompose=15, plan-to-prd=20, implement=75, fix=75, test=50, verify=100. Configurable from the Settings page.
 

package/docs/pr-review-fix-loop.md

@@ -34,7 +34,7 @@ How the engine manages the lifecycle of a PR from creation through review, fix,
 
 When multiple problems coexist, earlier triggers get the first chance to enqueue work. The local `fixDispatched` flag is declared before the fix triggers and set after human-feedback, review-feedback, and build-failure dispatches. Conflict fixes run last and explicitly require `!fixDispatched`, so any earlier successful fix dispatch suppresses the conflict fix for that PR in the same discovery pass. Build fixes are evaluated after human and minion review feedback, but the build-fix condition itself is not gated by `!fixDispatched`.
 
-`evalMaxIterations` only applies to the minion review loop: initial minion review, minion re-review, and minion review-feedback fixes. It does not gate human-feedback fixes, build-failure fixes, or merge-conflict fixes.
+The engine does not cap review→fix cycles or build-fix attempts. Each trigger evaluates its own gates on every discovery pass; loops stop only when the underlying condition clears (reviewer approves, build passes, conflict resolves, human feedback handled). Operators who need to halt automation on a runaway PR should disable the relevant feature flag (`evalLoop`, `autoFixBuilds`, `autoFixConflicts`) or close the PR.
 
 ### A. Human comments (`humanFeedback.pendingFix`)
 
@@ -42,16 +42,14 @@ When multiple problems coexist, earlier triggers get the first chance to enqueue
 - Agent comments filtered out via `/\bMinions\s*\(/i` regex on comment body
 - Coalesces multiple comments arriving during cooldown into single fix
 - Routes to author
-- Not gated by `_evalEscalated` — humans can always force more fixes via PR comments even after the minion review loop escalates.
 - Fix agents must treat human comments as claims to verify, not commands. They inspect or reproduce each claimed issue, make the smallest correct fix only when the claim is valid, and otherwise reply with evidence-backed rationale.
 
 ### B. Review feedback (`changes-requested`)
 
-- Gate: `reviewStatus === 'changes-requested'` + `!awaitingReReview` + `!evalEscalated` + not dispatched + not on cooldown
+- Gate: `reviewStatus === 'changes-requested'` + `!awaitingReReview` + not dispatched + not on cooldown
 - Routes to PR author via `_author_` routing token
 - `review_note` = reviewer's feedback
 - Sets `fixDispatched = true` — prevents the later conflict fix from also firing this pass
-- **Review-loop escalation**: after `evalMaxIterations` review→fix cycles (default 3), `_evalEscalated` is set on the PR and *only this trigger plus minion review/re-review* stop. Triggers A (human comments), C (build failures), and D (merge conflicts) keep running. The dashboard PR row distinguishes the two states with separate badges (review badge `review-escalated` vs. build badge `build-escalated`).
 - Fix agents validate each requested-change claim before editing. Invalid, stale, already-addressed, out-of-scope, or harmful feedback should get a respectful evidence-backed rebuttal rather than a success-shaped code change.
 
 ### C. Build failures (`buildStatus === 'failing'`)
@@ -59,8 +57,6 @@ When multiple problems coexist, earlier triggers get the first chance to enqueue
 - Gate: provider polling enabled (`adoPollEnabled` or `ghPollEnabled`) + `autoFixBuilds` + `buildStatus === 'failing'` + grace period expired
 - **Grace period** (`_buildFixPushedAt`): after fix dispatches, waits `buildFixGracePeriod` (default 10min, configurable in `ENGINE_DEFAULTS`) for CI to run before re-dispatching. Cleared when poller detects build status transition (CI actually ran).
 - **Error logs**: GitHub fetches annotations (failures only, not warnings) + Actions job log (always). ADO queries builds API directly (not status checks), fetches build timeline → failed task logs (up to 10 per build, up to 10 failing pipelines).
-- **Build-fix escalation**: after 3 failed attempts, writes an inbox alert, sets `buildFixEscalated = true`, and stops *only this trigger* (auto-dispatch for build fixes). The counter resets when the build recovers. Independent of `_evalEscalated`.
-- Not gated by `_evalEscalated` — build-fix is mechanical and runs even if the review loop has escalated.
 - Sets `fixDispatched = true` after dispatch so the later conflict trigger is suppressed in the same pass.
 
 ### D. Merge conflicts (`_mergeConflict`)
@@ -88,7 +84,7 @@ When multiple problems coexist, earlier triggers get the first chance to enqueue
 - Fix agent pushes → `_buildFixPushedAt` stamped
 - Poller detects new commit → CI starts → `buildStatus` transitions (`failing` → `running`)
 - `_buildFixPushedAt` cleared on any transition
-- If CI passes → `buildFixAttempts` reset, `buildErrorLog` cleared → done
+- If CI passes → `buildErrorLog` cleared → done
 - If CI fails again → fresh error logs fetched → new fix dispatches immediately (grace already cleared by transition)
 
 ## Race prevention
@@ -125,10 +121,6 @@ When multiple problems coexist, earlier triggers get the first chance to enqueue
 | `reviewStatus` | Poller + post-completion | `pending` / `approved` / `changes-requested` / `waiting` |
 | `buildStatus` | Poller | `none` / `passing` / `failing` / `running` |
 | `buildErrorLog` | Poller | Actual CI error output for fix agents |
-| `buildFixAttempts` | Discovery (on dispatch) | Counter for build-fix escalation cap |
-| `buildFixEscalated` | Discovery (on cap) | Stops *build-fix* auto-dispatch only (review/re-review and other fix triggers continue) |
-| `_reviewFixCycles` | Discovery (on dispatch) | Counter for review→fix cycle cap (`evalMaxIterations`) |
-| `_evalEscalated` | Discovery (on cap) | Stops *review/re-review and review-feedback fix* auto-dispatch only (build-fix, conflict-fix, and human-feedback fix continue). Cleared when reviewer eventually approves the PR. |
 | `_buildFixPushedAt` | Discovery (on dispatch) | Grace period timestamp |
 | `_buildFailNotified` | Discovery | Dedup for inbox alert |
 | `lastPushedAt` | Poller (new commit) | Tracks latest push for re-review logic |

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "@yemi33/minions",
-  "version": "0.1.1913",
+  "version": "0.1.1914",
   "description": "Multi-agent AI dev team that runs from ~/.minions/ — five autonomous agents share a single engine, dashboard, and knowledge base",
   "bin": {
     "minions": "bin/minions.js"