@mmerterden/multi-agent-pipeline 17.6.0 → 19.0.0

package/pipeline/schemas/agent-state.schema.json

@@ -21,7 +21,7 @@
   "properties": {
     "schemaVersion": {
       "type": "string",
-      "enum": ["2.0.0", "2.1.0"],
+      "enum": ["2.0.0", "2.1.0", "2.2.0"],
       "description": "v2.0.0: single-repo only (scalar project/projectRoot/worktreePath). v2.1.0: adds optional projects[] array for multi-repo tasks  -  single-repo fields remain for backward compat; when projects[] has >1 entries, multi-repo mode is active."
     },
     "taskId": {
@@ -193,7 +193,7 @@
     "maturity": {
       "type": "object",
       "additionalProperties": false,
-      "description": "How ready the fetched item was to be developed. Produced by lib/issue-fetcher.sh, which owns the gap codes and their localized wording. Declared as of v17.6.0 - it had been written by every issue-shaped run since long before and read by Phase 0, Phase 4 and the clarifier, while this schema forbade it under additionalProperties: false.",
+      "description": "How ready the fetched item was to be developed. Produced by lib/issue-fetcher.sh, which owns the gap codes and their localized wording. Declared as of v17.6.0 - it had been written by every issue-shaped run since long before and read by Phase 0, Phase 3 and the clarifier, while this schema forbade it under additionalProperties: false.",
       "properties": {
         "score": {
           "type": ["integer", "null"],
@@ -269,7 +269,7 @@
     "waitingFor": {
       "type": "string",
       "enum": ["maturity", "user-channels-choice"],
-      "description": "The step a paused run must RE-ENTER, as opposed to the phase it should continue past. Declared as of v17.6.0: Phase 7's channels pause has written it since long before, and resume/SKILL.md never read it, so that pause was resumable only in prose. resume reads this first and falls back to currentPhase + 1 when it is absent. maturity: Phase 0's maturity step, with the item re-fetched (refs/features/maturity-followup.md). user-channels-choice: Phase 7's channels menu, with the stored channelsInput."
+      "description": "The step a paused run must RE-ENTER, as opposed to the phase it should continue past. Declared as of v17.6.0: Phase 5's channels pause has written it since long before, and resume/SKILL.md never read it, so that pause was resumable only in prose. resume reads this first and falls back to currentPhase + 1 when it is absent. maturity: Phase 0's maturity step, with the item re-fetched (refs/features/maturity-followup.md). user-channels-choice: Phase 5's channels menu, with the stored channelsInput."
     },
     "workspaceSource": {
       "type": "string",
@@ -284,13 +284,13 @@
     "currentPhase": {
       "type": "integer",
       "minimum": 0,
-      "maximum": 7,
-      "description": "Last phase the orchestrator entered. Resume re-enters at currentPhase + 1."
+      "maximum": 5,
+      "description": "Last phase the orchestrator entered. Resume re-enters at currentPhase + 1. Six phases (0..5) since v19.0.0; a state written by an older version is migrated by schemas/migrations/state-2.1.0-to-2.2.0.mjs."
     },
     "status": {
       "type": "string",
       "enum": ["in_progress", "awaiting_input", "paused", "complete", "failed"],
-      "description": "Overall task status. `awaiting_input` is deliberately distinct from both `in_progress` and `paused`: the run is not broken and not resumable by a retry - it is finished with what it can do alone and is holding for a human answer, which is what Phase 7 does by design after the PR is open. Without a word for it, a run whose work had landed read exactly like one that crashed mid-development, and a state file in the wild had already invented `awaiting-user-test-main-checkout` to say it. `paused` stays what it was: stopped deliberately, resumable, nothing expected from anyone."
+      "description": "Overall task status. `awaiting_input` is deliberately distinct from both `in_progress` and `paused`: the run is not broken and not resumable by a retry - it is finished with what it can do alone and is holding for a human answer, which is what Phase 5 does by design after the PR is open. Without a word for it, a run whose work had landed read exactly like one that crashed mid-development, and a state file in the wild had already invented `awaiting-user-test-main-checkout` to say it. `paused` stays what it was: stopped deliberately, resumable, nothing expected from anyone."
     },
     "startedAt": {
       "type": "string",
@@ -307,7 +307,7 @@
     "relatedIssues": {
       "type": "array",
       "maxItems": 20,
-      "description": "Jira issues fetched alongside the task at intake: the other sub-tasks under this issue's parent, where a board keeps the analysis and the test scope. NOT the same field as siblings[] below, which is read-only sibling REPOS. Written by the picker bridge from descriptor.relatedIssues; read by Phase 1 as ground truth and by Phase 2's parent-story scope-drift check. Durable on purpose: /multi-agent:resume rebuilds context from artefacts, never from the conversation, so intake-only enrichment would vanish on the first resume. The task's own description and maturity are still NOT persisted here, which is a known asymmetry, not an oversight.",
+      "description": "Jira issues fetched alongside the task at intake: the other sub-tasks under this issue's parent, where a board keeps the analysis and the test scope. NOT the same field as siblings[] below, which is read-only sibling REPOS. Written by the picker bridge from descriptor.relatedIssues; read by Phase 1 as ground truth and by its own parent-story scope-drift check. Durable on purpose: /multi-agent:resume rebuilds context from artefacts, never from the conversation, so intake-only enrichment would vanish on the first resume. The task's own description and maturity are still NOT persisted here, which is a known asymmetry, not an oversight.",
       "items": {
         "type": "object",
         "additionalProperties": false,
@@ -342,7 +342,7 @@
     "siblings": {
       "type": "array",
       "maxItems": 10,
-      "description": "Repos the dev-context picker offered that this run does not modify: read-only siblings, plus any extra the user selected. Persisted at Phase 0 because the phases that consume them run much later - Phase 4's platform-parity cross-check reads this as the fourth of its four counterpart sources (after --with, prefs.projects[<slug>].counterpartRoots[] and the primary checkout's sibling directories; see multi-agent-refs/platform-parity.md), so a picker result that is not written here is a candidate the check can never see.",
+      "description": "Repos the dev-context picker offered that this run does not modify: read-only siblings, plus any extra the user selected. Persisted at Phase 0 because the phases that consume them run much later - Phase 3's platform-parity cross-check reads this as the fourth of its four counterpart sources (after --with, prefs.projects[<slug>].counterpartRoots[] and the primary checkout's sibling directories; see multi-agent-refs/platform-parity.md), so a picker result that is not written here is a candidate the check can never see.",
       "items": {
         "type": "object",
         "additionalProperties": false,
@@ -364,7 +364,7 @@
           "canPush": {
             "type": "boolean",
             "default": false,
-            "description": "Carried from the picker. Informational here: a sibling is read-only to Phase 4 regardless."
+            "description": "Carried from the picker. Informational here: a sibling is read-only to Phase 3 regardless."
           }
         }
       }
@@ -399,7 +399,7 @@
         "appliedPhase": {
           "type": ["integer", "null"],
           "minimum": 0,
-          "maximum": 7,
+          "maximum": 5,
           "description": "The phase that consumed it."
         }
       }
@@ -407,7 +407,7 @@
     "telemetry": {
       "type": "object",
       "additionalProperties": true,
-      "description": "Run telemetry. mcpCalls[] is consumed by smoke-no-mcp-in-dev-phases.sh to enforce the 'no MCP outside analysis' contract  -  every mcp__* invocation MUST append an entry tagged with the phase it ran in. skillCalls[] is Phase 3's self-report of which skills, plugin skills and guides it consulted; Phase 4 Step 1.78 reads it as CORROBORATING evidence only and never computes coverage from it.",
+      "description": "Run telemetry. mcpCalls[] is consumed by smoke-no-mcp-in-dev-phases.sh to enforce the 'no MCP outside analysis' contract  -  every mcp__* invocation MUST append an entry tagged with the phase it ran in. skillCalls[] is Phase 3's self-report of which skills, plugin skills and guides it consulted; Phase 3 Step 1.78 reads it as CORROBORATING evidence only and never computes coverage from it.",
       "properties": {
         "mcpCalls": {
           "type": "array",
@@ -423,7 +423,7 @@
               "phase": {
                 "type": "integer",
                 "minimum": 0,
-                "maximum": 7,
+                "maximum": 5,
                 "description": "Pipeline phase the call ran in. Only 0 (init) and 1 (analysis) are permitted."
               },
               "timestamp": {
@@ -449,7 +449,7 @@
               "phase": {
                 "type": "integer",
                 "minimum": 0,
-                "maximum": 7,
+                "maximum": 5,
                 "description": "Phase the consultation happened in. Normally 3."
               },
               "targetFiles": {
@@ -465,7 +465,7 @@
               },
               "routedBy": {
                 "type": "string",
-                "description": "Set when a stack toolkit's own index skill chose this skill, as '<toolkit>:index@<version>'. Recorded so a finding can be traced to the index version that selected it  -  the skill set differs between plugin versions. Phase 4 Step 1.78 surfaces these separately in the manifest ledger; it does NOT grant them extra trust, because the resolver stays primary either way."
+                "description": "Set when a stack toolkit's own index skill chose this skill, as '<toolkit>:index@<version>'. Recorded so a finding can be traced to the index version that selected it  -  the skill set differs between plugin versions. Phase 3 Step 1.78 surfaces these separately in the manifest ledger; it does NOT grant them extra trust, because the resolver stays primary either way."
               }
             }
           }
@@ -479,7 +479,7 @@
     "onlyDevelop": {
       "type": "boolean",
       "default": false,
-      "description": "Short pipeline  -  phases 1 and 2 are skipped. Set by the Phase 0 Step 7.5 depth picker as of v16.0.0; the key and every reader of it are unchanged. Phase 4 Review still runs (v14.0.0+). Always false in an autopilot run, which never asks the depth question."
+      "description": "Short pipeline  -  phases 1 and 2 are skipped. Set by the Phase 0 Step 7.5 depth picker as of v16.0.0; the key and every reader of it are unchanged. Phase 3 Review still runs (v14.0.0+). Always false in an autopilot run, which never asks the depth question."
     },
     "localMode": {
       "type": "boolean",
@@ -526,11 +526,11 @@
       "properties": {
         "lastAnalysisDigest": {
           "type": ["string", "null"],
-          "description": "The evidence_digest the analysis documents were written from, persisted by Phase 1. Phase 3 compares it against the document front-matter; absent means the freshness check is not-verifiable, never fresh."
+          "description": "The evidence_digest the analysis documents were written from, persisted by Phase 1. Phase 2 compares it against the document front-matter; absent means the freshness check is not-verifiable, never fresh."
         },
         "analysisBaseCommit": {
           "type": ["string", "null"],
-          "description": "git rev-parse HEAD at analysis emit time. Phase 3 diffs it against HEAD to detect repo drift under a reused spec, which the digest alone cannot see."
+          "description": "git rev-parse HEAD at analysis emit time. Phase 2 diffs it against HEAD to detect repo drift under a reused spec, which the digest alone cannot see."
         }
       }
     },
@@ -553,14 +553,14 @@
         "reviewBlocking": {
           "type": "boolean",
           "default": false,
-          "description": "Set with tier 3. Phase 4 holds the run at review_blocking until a human signs off on the component choice."
+          "description": "Set with tier 3. Phase 3 holds the run at review_blocking until a human signs off on the component choice."
         }
       }
     },
     "designCheck": {
       "type": ["object", "null"],
       "additionalProperties": true,
-      "description": "State for a /multi-agent:design-check run. Populated in Phase 0 (mock feasibility, scenario inventory, scope), Phase 2 (Figma variants) and Phase 3 (captures, skips, findings).",
+      "description": "State for a /multi-agent:design-check run. Populated in Phase 0 (mock feasibility, scenario inventory, scope), Phase 1 (Figma variants) and Phase 2 (captures, skips, findings).",
       "properties": {
         "module": {
           "type": "string",
@@ -595,7 +595,7 @@
         "inventory": {
           "type": ["object", "null"],
           "additionalProperties": true,
-          "description": "design_scenario_inventory result, persisted verbatim. THIS is the audit's target set  -  Phase 3 iterates it and Phase 4 gates on it. Never re-derive it by reading the repo.",
+          "description": "design_scenario_inventory result, persisted verbatim. THIS is the audit's target set  -  Phase 2 iterates it and Phase 3 gates on it. Never re-derive it by reading the repo.",
           "properties": {
             "targetCount": {
               "type": "integer"
@@ -666,7 +666,7 @@
         },
         "skipped": {
           "type": "array",
-          "description": "Targets that could not be audited, each with a CONCRETE reason. 'Requires a scenario / prefix / launch-arg' is not a reason  -  that is the work Phase 3 exists to do. A skip without a reason fails the coverage gate.",
+          "description": "Targets that could not be audited, each with a CONCRETE reason. 'Requires a scenario / prefix / launch-arg' is not a reason  -  that is the work Phase 2 exists to do. A skip without a reason fails the coverage gate.",
           "items": {
             "type": "object",
             "additionalProperties": true,
@@ -743,9 +743,9 @@
     },
     "phases": {
       "type": "object",
-      "description": "Per-phase status + outputs. Keys are phase numbers as strings (\"0\"..\"7\").",
+      "description": "Per-phase status + outputs. Keys are phase numbers as strings (\"0\"..\"5\").",
       "patternProperties": {
-        "^[0-7]$": {
+        "^[0-5]$": {
           "type": "object",
           "additionalProperties": false,
           "properties": {
@@ -778,7 +778,7 @@
             },
             "subStep": {
               "type": "string",
-              "description": "Sub-step checkpoint for long phases (3, 7) so resume re-does only the unfinished tail instead of the whole phase. Short token, e.g. 'red'|'green'|'build'|'pr-opened'|'confluence-synced'. See operations.md 'Sub-step checkpoints'."
+              "description": "Sub-step checkpoint for long phases (2, 5) so resume re-does only the unfinished tail instead of the whole phase. Short token, e.g. 'red'|'green'|'build'|'pr-opened'|'confluence-synced'. See operations.md 'Sub-step checkpoints'."
             },
             "notes": {
               "type": "string"
@@ -787,38 +787,38 @@
               "type": "integer",
               "minimum": 0,
               "maximum": 2,
-              "description": "v5.3.0 Phase 2 Plan Approval Gate  -  number of clarification question/answer rounds the orchestrator ran before producing the first plan. Cap 2; hitting the cap produces a 'best-effort' plan instead of asking more questions. Only emitted on a Full interactive run (a Short run has no plan, autopilot may not ask)."
+              "description": "v5.3.0 Phase 1 Plan Approval Gate  -  number of clarification question/answer rounds the orchestrator ran before producing the first plan. Cap 2; hitting the cap produces a 'best-effort' plan instead of asking more questions. Only emitted on a Full interactive run (a Short run has no plan, autopilot may not ask)."
             },
             "clarificationQuestions": {
               "type": "array",
               "items": {
                 "type": "string"
               },
-              "description": "v5.3.0 Phase 2  -  ordered log of questions the orchestrator asked the user when the issue description was ambiguous (missing acceptance criteria, vague language, missing Figma/endpoint links, unclear scope vs. parent story)."
+              "description": "v5.3.0 Phase 1  -  ordered log of questions the orchestrator asked the user when the issue description was ambiguous (missing acceptance criteria, vague language, missing Figma/endpoint links, unclear scope vs. parent story)."
             },
             "clarificationAnswers": {
               "type": "array",
               "items": {
                 "type": "string"
               },
-              "description": "v5.3.0 Phase 2  -  ordered log of user answers aligned with clarificationQuestions (same length)."
+              "description": "v5.3.0 Phase 1  -  ordered log of user answers aligned with clarificationQuestions (same length)."
             },
             "planIterations": {
               "type": "integer",
               "minimum": 1,
-              "description": "v5.3.0 Phase 2 Plan Approval Gate  -  how many plan revisions the user requested before approving. 1 = approved on first show; N = user sent N-1 free-text edit requests and the plan was re-rendered each time."
+              "description": "v5.3.0 Phase 1 Plan Approval Gate  -  how many plan revisions the user requested before approving. 1 = approved on first show; N = user sent N-1 free-text edit requests and the plan was re-rendered each time."
             },
             "planApprovedAt": {
               "type": ["string", "null"],
               "format": "date-time",
-              "description": "v5.3.0 Phase 2 Plan Approval Gate  -  timestamp when the user approved the plan. Null when the gate is not applicable (Short run, or autopilot) or the task was aborted at the gate."
+              "description": "v5.3.0 Phase 1 Plan Approval Gate  -  timestamp when the user approved the plan. Null when the gate is not applicable (Short run, or autopilot) or the task was aborted at the gate."
             },
             "planEditRequests": {
               "type": "array",
               "items": {
                 "type": "string"
               },
-              "description": "v5.3.0 Phase 2  -  free-text edit instructions the user typed between plan renders. Preserved verbatim for audit; the planning model (Fable top tier) parses them conversationally to revise the plan."
+              "description": "v5.3.0 Phase 1  -  free-text edit instructions the user typed between plan renders. Preserved verbatim for audit; the planning model (Fable top tier) parses them conversationally to revise the plan."
             }
           }
         }
@@ -827,7 +827,7 @@
     "baseline": {
       "type": "object",
       "additionalProperties": true,
-      "description": "Pre-work state of the repo, captured in Phase 0. Exists so Phase 4 can tell an inherited failure from one this run caused.",
+      "description": "Pre-work state of the repo, captured in Phase 0. Exists so Phase 3 can tell an inherited failure from one this run caused.",
       "properties": {
         "tests": {
           "type": "object",
@@ -844,7 +844,7 @@
               "items": {
                 "type": "string"
               },
-              "description": "Identifiers of tests already failing before this run. Empty on a red status means the output could not be parsed into names: the evidence is then the logPath alone, and Phase 4 reports 'inherited red, not attributable' rather than inventing a set."
+              "description": "Identifiers of tests already failing before this run. Empty on a red status means the output could not be parsed into names: the evidence is then the logPath alone, and Phase 3 reports 'inherited red, not attributable' rather than inventing a set."
             },
             "logPath": {
               "type": "string",
@@ -856,7 +856,7 @@
             },
             "command": {
               "type": "string",
-              "description": "The exact test command that produced this baseline. Phase 4 compares against its own Gate 3 command and treats a mismatch as unknown."
+              "description": "The exact test command that produced this baseline. Phase 3 compares against its own Gate 3 command and treats a mismatch as unknown."
             }
           }
         }
@@ -866,7 +866,7 @@
       "type": "array",
       "items": {
         "type": "object",
-        "description": "Per-iteration review record. It accretes fields across phases - summary counts in Phase 4, plus the merged reviewers[] and triage{} detail and validatorResult that run-metrics.mjs and Phase 7 read - so extra keys are allowed.",
+        "description": "Per-iteration review record. It accretes fields across phases - summary counts in Phase 4, plus the merged reviewers[] and triage{} detail and validatorResult that run-metrics.mjs and Phase 5 read - so extra keys are allowed.",
         "additionalProperties": true,
         "properties": {
           "iteration": {
@@ -892,7 +892,7 @@
           "delta": {
             "type": "object",
             "additionalProperties": true,
-            "description": "Written by Phase 4 Step 3.8 from review-delta.mjs: how this round's accepted findings relate to the previous round's rework mandate. Absent on iteration 1 and on runs from before the delta existed.",
+            "description": "Written by Phase 3 Step 3.8 from review-delta.mjs: how this round's accepted findings relate to the previous round's rework mandate. Absent on iteration 1 and on runs from before the delta existed.",
             "properties": {
               "previousIteration": {
                 "type": "integer",
@@ -1050,7 +1050,7 @@
           "anonymizationMap": {
             "type": "object",
             "additionalProperties": true,
-            "description": "Written by Phase 4 Step 3.0 from anonymize-findings.mjs --map, read by run-metrics.mjs to attribute accepted findings back to a reviewer. Declared because the phase doc names it and a consumer reads it; absent on runs from before anonymization existed, which run-metrics reports as perReviewerAttribution \"unavailable\" rather than as a zero. Never goes into a prompt: it is the mapping the anonymization exists to withhold.",
+            "description": "Written by Phase 3 Step 3.0 from anonymize-findings.mjs --map, read by run-metrics.mjs to attribute accepted findings back to a reviewer. Declared because the phase doc names it and a consumer reads it; absent on runs from before anonymization existed, which run-metrics reports as perReviewerAttribution \"unavailable\" rather than as a zero. Never goes into a prompt: it is the mapping the anonymization exists to withhold.",
             "properties": {
               "seed": {
                 "type": "string",
@@ -1078,7 +1078,7 @@
     "circuitBreaker": {
       "type": "object",
       "additionalProperties": false,
-      "description": "Autopilot circuit-breaker record (refs/features/autopilot-circuit-breaker.md). Written only when a trigger trips: trigger 2 by Phase 4 Step 3.8 (a mandate finding survived identicalFindingCycles rework cycles), trigger 3 by the Phase 3 re-entry hard-kill, trigger 6 by Phase 0 Step 3 when autopilot posted the base-branch question on the issue and must not answer it itself. /multi-agent:resume clears tripped and keeps counters.",
+      "description": "Autopilot circuit-breaker record (refs/features/autopilot-circuit-breaker.md). Written only when a trigger trips: trigger 2 by Phase 3 Step 3.8 (a mandate finding survived identicalFindingCycles rework cycles), trigger 3 by the Phase 3 re-entry hard-kill, trigger 6 by Phase 0 Step 3 when autopilot posted the base-branch question on the issue and must not answer it itself. /multi-agent:resume clears tripped and keeps counters.",
       "required": ["tripped"],
       "properties": {
         "tripped": {
@@ -1099,7 +1099,7 @@
             "phase": {
               "type": "integer",
               "minimum": 0,
-              "maximum": 7
+              "maximum": 5
             },
             "step": {
               "type": "string"
@@ -1133,7 +1133,7 @@
     "diffRisk": {
       "type": "object",
       "additionalProperties": true,
-      "description": "Totals from diff-risk-score.mjs, persisted by Phase 4 Step 1.75 so Phase 6 (PR risk section), Phase 7 and run-metrics.mjs read the same numbers the review scope decision used.",
+      "description": "Totals from diff-risk-score.mjs, persisted by Phase 3 Step 1.75 so Phase 4 (PR risk section), Phase 5 and run-metrics.mjs read the same numbers the review scope decision used.",
       "properties": {
         "files": {
           "type": "integer",
@@ -1295,7 +1295,7 @@
             "description": "v2.1.0+ push-must-succeed retry counter. Increments per rebase-retry."
           },
           "buildStatus": {
-            "description": "Latest Phase 3 build outcome for this repo. A string in early phases ('pending'/'green'/'passed'/'failed') or null before any attempt; once Phase 3 runs a build it becomes a {ok, attempts, lastError} object (see phase-3-dev.md). run-metrics.mjs reads both the string and object forms.",
+            "description": "Latest Phase 3 build outcome for this repo. A string in early phases ('pending'/'green'/'passed'/'failed') or null before any attempt; once Phase 3 runs a build it becomes a {ok, attempts, lastError} object (see phase-2-dev.md). run-metrics.mjs reads both the string and object forms.",
             "anyOf": [
               {
                 "type": "string",
@@ -1327,16 +1327,16 @@
     "worktreeRemovedAt": {
       "type": ["string", "null"],
       "format": "date-time",
-      "description": "Set when Phase 6 removed the worktree after opening the PR. Its presence is what tells :resume, :status and :log that a worktree-less task is finished-and-tidied rather than broken  -  without it a missing worktree is indistinguishable from a killed run."
+      "description": "Set when Phase 4 removed the worktree after opening the PR. Its presence is what tells :resume, :status and :log that a worktree-less task is finished-and-tidied rather than broken  -  without it a missing worktree is indistinguishable from a killed run."
     },
     "artifactsPath": {
       "type": ["string", "null"],
-      "description": "Directory the worktree's artefacts were salvaged into before removal (agent-state, phase-tracker, triage-output, .pipeline/, build+test logs, review diff). Phase 7 and :resume read from here when worktreePath is gone."
+      "description": "Directory the worktree's artefacts were salvaged into before removal (agent-state, phase-tracker, triage-output, .pipeline/, build+test logs, review diff). Phase 5 and :resume read from here when worktreePath is gone."
     },
     "testDepth": {
       "type": ["string", "null"],
       "enum": ["unit", "unit+ui", "unit+mcp", null],
-      "description": "How far the run tests, answered at intake because Phase 5 is absent from four of the eight modes and a question asked where it cannot be reached is a question nobody answers. `unit+ui` runs the repo's own UI test and records the screen around it; `unit+mcp` drives the flow through the toolkit MCP instead. The options offered are built from evidenceCapability, never from the model's reading of the repo."
+      "description": "How far the run tests, answered at intake because the user test is absent from four of the six modes and a question asked where it cannot be reached is a question nobody answers. `unit+ui` runs the repo's own UI test and records the screen around it; `unit+mcp` drives the flow through the toolkit MCP instead. The options offered are built from evidenceCapability, never from the model's reading of the repo."
     },
     "testDepthSource": {
       "type": ["string", "null"],
@@ -1361,7 +1361,7 @@
       "items": {
         "type": "string"
       },
-      "description": "The LANGUAGE census (ios, python, node, go, docker, monorepo), written by Phase 1 Step 2 and read by graph-build.mjs --stack, the dev-critic checklist selector and the Phase 4 reviewer skill tables. Deliberately not narrowed to the stacks[] vocabulary: graph-build accepts ios|android|node|python|go and has no notion of web or backend. Phase docs have instructed this write since before the schema existed, and additionalProperties is false, so it is declared here rather than silently invalid."
+      "description": "The LANGUAGE census (ios, python, node, go, docker, monorepo), written by Phase 1 Step 2 and read by graph-build.mjs --stack, the dev-critic checklist selector and the Phase 3 reviewer skill tables. Deliberately not narrowed to the stacks[] vocabulary: graph-build accepts ios|android|node|python|go and has no notion of web or backend. Phase docs have instructed this write since before the schema existed, and additionalProperties is false, so it is declared here rather than silently invalid."
     },
     "evidenceCapability": {
       "type": ["object", "null"],
@@ -1459,7 +1459,7 @@
     "visualEvidence": {
       "type": ["object", "null"],
       "additionalProperties": false,
-      "description": "Before/after screenshots and the UI flow video for a UI change, attached to the Jira issue and rendered in the Phase 7 comment and the PR body. Required decided mechanically from taskType plus the changed-file list. Contract: multi-agent-refs/features/visual-evidence.md.",
+      "description": "Before/after screenshots and the UI flow video for a UI change, attached to the Jira issue and rendered in the Phase 5 comment and the PR body. Required decided mechanically from taskType plus the changed-file list. Contract: multi-agent-refs/features/visual-evidence.md.",
       "properties": {
         "required": {
           "type": "boolean"
@@ -1530,21 +1530,21 @@
               }
             }
           },
-          "description": "Captured in Phase 3 after the build+test gate, because Phase 5 is dropped by every autopilot and --local entry."
+          "description": "Captured in Phase 2 after the build+test gate, because the user test is dropped by every autopilot and --local entry."
         },
         "host": {
           "type": ["string", "null"],
           "enum": ["jira", "github-public", "github-private", "none", null],
-          "description": "Where the artefacts are published, resolved in Phase 6. `jira` attaches both stills and video. `github-public` pushes the stills to the evidence branch and embeds them in the PR body. `github-private` pushes the same stills but the PR carries a blob permalink instead of an inline image, because GitHub's image proxy cannot fetch a private repo's raw URL and an embedded one renders broken for every reader. `none` publishes nothing and records the gap. Video is Jira-only by decision: without an attachment host there is nothing a recording can be attached to."
+          "description": "Where the artefacts are published, resolved in Phase 4. `jira` attaches both stills and video. `github-public` pushes the stills to the evidence branch and embeds them in the PR body. `github-private` pushes the same stills but the PR carries a blob permalink instead of an inline image, because GitHub's image proxy cannot fetch a private repo's raw URL and an embedded one renders broken for every reader. `none` publishes nothing and records the gap. Video is Jira-only by decision: without an attachment host there is nothing a recording can be attached to."
         },
         "hostReason": {
           "type": ["string", "null"],
-          "description": "Why this host and not the one above it in the order. A host of `none` with no reason is the silence the Phase 6 blocker exists to catch."
+          "description": "Why this host and not the one above it in the order. A host of `none` with no reason is the silence the Phase 4 blocker exists to catch."
         },
         "videoTier": {
           "type": ["integer", "null"],
           "enum": [1, 2, 3, null],
-          "description": "1 = the repo's own UI test target drove the flow, 2 = MCP-driven flow, 3 = no recording. Resolved from the capability probe, then RE-CHECKED at capture time: a device booted at intake can be gone by Phase 3, and a tier recorded from a stale measurement is a promise the run cannot keep."
+          "description": "1 = the repo's own UI test target drove the flow, 2 = MCP-driven flow, 3 = no recording. Resolved from the capability probe, then RE-CHECKED at capture time: a device booted at intake can be gone by Phase 2, and a tier recorded from a stale measurement is a promise the run cannot keep."
         },
         "videoTierReason": {
           "type": ["string", "null"],
@@ -1582,7 +1582,7 @@
         },
         "gaps": {
           "type": "array",
-          "description": "Every artefact that is required and absent, with its reason. A recorded reason satisfies the Phase 6 blocker; silence does not.",
+          "description": "Every artefact that is required and absent, with its reason. A recorded reason satisfies the Phase 4 blocker; silence does not.",
           "items": {
             "type": "object",
             "additionalProperties": false,

package/pipeline/schemas/analysis-output.schema.json

@@ -3,7 +3,7 @@
   "$id": "https://github.com/mmerterden/multi-agent-pipeline/pipeline/schemas/analysis-output.schema.json",
   "version": "1.0.0",
   "title": "Multi-Agent Pipeline  -  Phase 1 analysis output",
-  "description": "Contract for the Explorer subagent's summary in Phase 1. Drives Phase 2 task decomposition and Phase 4 review focus. Explorer must not mutate the codebase; output is read-only research.",
+  "description": "Contract for the Explorer subagent's summary in Phase 1. Drives Phase 2 task decomposition and Phase 3 review focus. Explorer must not mutate the codebase; output is read-only research.",
   "type": "object",
   "additionalProperties": false,
   "required": ["stack", "touchedAreas", "risks", "summary"],
@@ -67,7 +67,7 @@
     "summary": {
       "type": "string",
       "minLength": 20,
-      "description": "≤200 word plain-language summary for Phase 2 planning + user approval."
+      "description": "≤200 word plain-language summary for Phase 1 planning + user approval."
     },
     "docStatus": {
       "type": "string",

package/pipeline/schemas/autopilot-config.schema.json

@@ -114,7 +114,7 @@
     "channels": {
       "type": "object",
       "additionalProperties": false,
-      "description": "Phase 7 pauses for channel selection in every attended mode, by design. Unattended there is nobody to ask, so the answer comes from here instead and Phase 7 completes without prompting. Without this the queue would generate its own backlog: every finished item parked waiting for a menu.",
+      "description": "Phase 5 pauses for channel selection in every attended mode, by design. Unattended there is nobody to ask, so the answer comes from here instead and Phase 5 completes without prompting. Without this the queue would generate its own backlog: every finished item parked waiting for a menu.",
       "properties": {
         "mode": {
           "enum": ["none", "pr-body", "configured"],

package/pipeline/schemas/code-graph.schema.json

@@ -2,7 +2,7 @@
   "$schema": "https://json-schema.org/draft/2020-12/schema",
   "$id": "https://github.com/mmerterden/multi-agent-pipeline/schemas/code-graph.schema.json",
   "title": "Code graph",
-  "description": "Deterministic, LLM-free code graph produced by graph-build.mjs and consumed by Phase 1 (Explore scope narrowing) and Phase 7 (knowledge capture). One file per project under ~/.claude/knowledge/<project>/code-graph.json.",
+  "description": "Deterministic, LLM-free code graph produced by graph-build.mjs and consumed by Phase 1 (Explore scope narrowing) and Phase 5 (knowledge capture). One file per project under ~/.claude/knowledge/<project>/code-graph.json.",
   "type": "object",
   "required": ["schemaVersion", "stack", "root", "generatedAt", "stats", "nodes", "edges"],
   "additionalProperties": false,

package/pipeline/schemas/criteria-manifest.schema.json

@@ -2,7 +2,7 @@
   "$schema": "https://json-schema.org/draft/2020-12/schema",
   "$id": "https://github.com/mmerterden/multi-agent-pipeline/pipeline/schemas/criteria-manifest.schema.json",
   "title": "Criteria manifest",
-  "description": "What the changed code was supposed to honour, resolved deterministically by skill-conformance.mjs BEFORE any reviewer runs. Produced in Phase 4 Step 1.78, consumed by the reviewer prompt assembly (the ${CRITERIA} block) and by the triage merge. Written to disk so the denominator of 'was this done completely' is fixed in advance and cannot be renegotiated by a reviewer after the fact.",
+  "description": "What the changed code was supposed to honour, resolved deterministically by skill-conformance.mjs BEFORE any reviewer runs. Produced in Phase 3 Step 1.78, consumed by the reviewer prompt assembly (the ${CRITERIA} block) and by the triage merge. Written to disk so the denominator of 'was this done completely' is fixed in advance and cannot be renegotiated by a reviewer after the fact.",
   "type": "object",
   "additionalProperties": false,
   "required": [

package/pipeline/schemas/dev-critic-output.schema.json

@@ -2,7 +2,7 @@
   "$schema": "http://json-schema.org/draft-07/schema#",
   "$id": "https://example.com/pipeline/dev-critic-output.schema.json",
   "title": "Dev Critic Output",
-  "description": "Schema for Phase 3.5 evaluator-optimizer (dev-critic) sub-agent output. Source: pipeline/agents/dev-critic.md. Pattern reference: Anthropic 'Building Effective Agents' (Dec 2024)  -  evaluator-optimizer.",
+  "description": "Schema for Phase 2.5 evaluator-optimizer (dev-critic) sub-agent output. Source: pipeline/agents/dev-critic.md. Pattern reference: Anthropic 'Building Effective Agents' (Dec 2024)  -  evaluator-optimizer.",
   "type": "object",
   "additionalProperties": false,
   "required": ["pass", "iteration", "gates", "findings"],

package/pipeline/schemas/diff-risk.schema.json

@@ -3,7 +3,7 @@
   "$id": "https://github.com/mmerterden/multi-agent-pipeline/pipeline/schemas/diff-risk.schema.json",
   "version": "1.1.0",
   "title": "Multi-Agent Pipeline  -  Phase 4 diff risk score",
-  "description": "Output contract for diff-risk-score.mjs. Heuristic, deterministic, no LLM. Produced before Phase 4 Step 2 to give reviewer prompts a priority ordering  -  never used as a gate. v1.1.0 adds the test_lines_removed signal (immutable-test backstop: a test file whose diff removes more lines than it adds).",
+  "description": "Output contract for diff-risk-score.mjs. Heuristic, deterministic, no LLM. Produced before Phase 3 Step 2 to give reviewer prompts a priority ordering  -  never used as a gate. v1.1.0 adds the test_lines_removed signal (immutable-test backstop: a test file whose diff removes more lines than it adds).",
   "type": "object",
   "additionalProperties": false,
   "required": ["schemaVersion", "task", "totals", "files"],

package/pipeline/schemas/migrations/prefs-2.4.0-to-2.5.0.mjs

@@ -2,7 +2,7 @@
  * prefs-2.4.0-to-2.5.0.mjs  -  preference schema upgrade 2.4.0 -> 2.5.0
  *
  * v14.0.0 put Phase 4 Review into the four `--dev` mode phase sets and added the
- * criteria-resolution stage (Phase 4 Step 1.78). Two preference keys follow:
+ * criteria-resolution stage (Phase 3 Step 1.78). Two preference keys follow:
  *
  *   - global.skillConformance.blockOnCoverageGap  (default FALSE)
  *       Whether a declared coverage gap  -  a language in the diff that no rule
@@ -13,7 +13,7 @@
  *       either way; this only controls whether it stops the pipeline.
  *
  *       There is deliberately no key for the stage itself, and none for the
- *       exception-expiry check, on the same grounds as Phase 4 Step 1.76: a run
+ *       exception-expiry check, on the same grounds as Phase 3 Step 1.76: a run
  *       that can switch off its own anti-reward-hacking control cannot be
  *       trusted to report a pass.
  *

package/pipeline/schemas/migrations/prefs-2.6.0-to-2.7.0.mjs

@@ -0,0 +1,31 @@
+/**
+ * prefs-2.6.0-to-2.7.0.mjs  -  preferences migration
+ *
+ * v19.0.0 removed Lite mode from the analysis phase, so
+ * `global.analysisPhase.mode` no longer accepts `auto` or `lite`.
+ *
+ * Lite chose sections from a fixed list (1, 2, 4, 9, 13, 14, 21 plus optional
+ * 23) selected by a three-signal score, while Locked 2 chooses them from
+ * evidence. The two disagreed in both directions: a small feature with rich
+ * business rules lost Section 15 because it was not on the list, and a feature
+ * with no API contract kept Section 9 because it was. Evidence decides now.
+ *
+ * The key is kept rather than deleted. A user who never set it sees nothing;
+ * a user who did keeps a valid file instead of one the schema rejects on the
+ * next run.
+ *
+ * @param {object} data  -  parsed multi-agent-preferences.json
+ * @returns {object}     -  migrated data with schemaVersion "2.7.0"
+ */
+export default function migrate(data) {
+  const out = JSON.parse(JSON.stringify(data));
+
+  out.schemaVersion = "2.7.0";
+
+  const ap = out?.global?.analysisPhase;
+  if (ap && typeof ap === "object" && typeof ap.mode === "string" && ap.mode !== "full") {
+    ap.mode = "full";
+  }
+
+  return out;
+}