verify-against-rubric-fleet

verify-against-rubric-fleet

Act as a senior/staff engineering orchestrator dispatching N stateless rubric-verifier sub-agents (one per artifact) and synthesizing their independent verdicts into a single VERIFICATION_LOG.md cohort entry with failure clustering.

Goal: For cohorts of N >= 4 artifacts that share ONE locked rubric (multi-slice retrospective, batch PR package review, design system spec audit), dispatch N stateless judgment workers in parallel. Each worker's evaluation context is ONLY its assigned artifact + the rubric; the artifact ID, output schema and selected-carrier envelope supply delivery metadata (no TASK_STATE.md, no DECISIONS.md, no sibling artifacts, no prior conversation). Workers return structured per-criterion verdicts independently. The orchestrator merges into one VERIFICATION_LOG.md cohort entry with: per-artifact verdicts, aggregate pass/fail/needs_revision counts, AND failure clustering -- which criteria fail across multiple artifacts (signal: likely a system-level issue not an individual artifact issue). Closes the same-context bias gap at cohort scale (Anthropic Outcomes 2026-05-06; the +10pp uplift compounds across N artifacts since each verification is independent).

Mandatory context bootstrap (before any output):

• Read these sections in WORKFLOW_OPERATING_SYSTEM.md first:
  • ## LLM execution contract
  • ## Editor mode policy (mode definitions only; the tool mapping table is lazy-loaded in wos/editor-mode-mappings.md and needed only for non-Claude-Code tools)
  • ## Global output contract (including Adaptive handoff and Mode selection rule)
  • ## Cross-cutting workflow guardrails
• Bootstrap tiers: the light-weight commands (branch-commit, what-next, where-we-at, slice-closure, compact-task-memory) plus the high-frequency implement-approved-slice and sync-task-state (v3 wave1 item D) read the four sections above with two subsections of ## Cross-cutting workflow guardrails skipped: ### External web access (centralized) and ### Sequencing heuristics (by phase). Everything else is read at every tier, including ### Proposal vs approved persistence and ### Substrate peer ownership (per ADR-0034), since all seven write substrate sections and reason about PROPOSED (state-reconcile stays on the full tier for cross-artifact judgment). The full tier is measured at 12109 tokens, the four always-read sections combined; the two skipped subsections are 1,034 of those (measured 2026-09-28), so the reduced tier is about 11,074. The leaf-reviewer tier (verify-against-rubric, ADR-0226) reads only ## Global output contract, measured at 5215 tokens, plus its rubric.
• Session bootstrap reuse (skip-if-unchanged; v3 wave1 item D): WHEN this conversation already read the bootstrap sections in an earlier turn still VISIBLE in the context window AND WORKFLOW_OPERATING_SYSTEM.md has not changed since, the command MAY skip the re-read and cite the earlier one, emitting one Command transcript line: Bootstrap: reusing turn <N> read, WOS unchanged. Scoped exception to the context-budget re-fetch rule (wos/context-budget.md, "The re-fetch rule"), because these sections are one large, static, byte-identical read repeated every turn; every other tool result still re-fetches. VISIBLE means the section text itself is still present and quotable now, not merely that a record of the earlier read exists: a harness that clears a tool result while the record survives (ADR-0114) has not satisfied VISIBLE, and self-declared memory after a compaction never qualifies. A stateless-per-turn harness is excluded. The transcript line is mandatory; a silent skip is invalid output.
• Resolving the spec and wos/<topic>.md: canonical repository FIRST. Otherwise use active-adapter docs: ~/.claude/workflow-docs (CLAUDE_WORKFLOW_DOCS_DIR), ~/.cursor/workflow-docs (WORKFLOW_DOCS_DIR; project .cursor/workflow-docs on project installs), ~/.codex/workflow-docs (CODEX_WORKFLOW_DOCS_DIR), or <kimi-home>/workflow-docs (KIMI_WORKFLOW_DOCS_DIR). Nonempty overrides win. Kimi home: KIMI_CODE_HOME, else existing ~/.kimi-code, else existing ~/.kimi, default ~/.kimi-code. Use one root for both paths; try other adapters if absent or unknown. Report resolved root, fallback to other adapters, or missing mandatory context in ### Command transcript. Installed snapshots refresh and prune retired topics only on sync.
• Read additional sections only when relevant to this command's role.
• Align all routing recommendations and next-command suggestions with the current command set.
• Official next-command names only: recommendations, including Run now, use registered names from commands/<name>.md or commands/<name>/SKILL.md; installed mode uses the available Fhorja command or skill inventory. No invented aliases. Run now: none with Mode: N/A ends only if no honest next step exists, defined under ### Official command names (routing integrity) (ADR-0126). Required inputs:
• task folder path
• artifact manifest: array of {artifact_id, artifact_path} per artifact (caller pre-enumerates; orchestrator does NOT auto-discover artifacts)
• rubric source: EITHER inline rubric (verbatim) OR reference to a section in IMPLEMENTATION_PLAN.md or DECISIONS.md (path + section anchor)
• optional: explicit criteria list (overrides the criteria extracted from rubric; useful when rubric prose mixes criteria with prologue)
• optional: explicit max_fanout override, lower only (defaults to 9, the review and verification cap in WORKFLOW_OPERATING_SYSTEM.md ## Parallel workflow, ADR-0236; never above 9 here)
• optional: cohort_label (string used in VERIFICATION_LOG.md entry header; defaults to date + N artifacts)