Validate process decomposition and architecture design quality before execution begins. Load when the setup-evaluator agent fires (automatic for agent-chain tasks), or when user says "evaluate this setup", "check the decomposition", "validate the architecture", "is this plan sound", "review the agent design". Catches structural errors, missing knowledge, unrealistic step ordering, and topology mismatches. Does NOT modify — only evaluates.
التثبيت
التثبيت باستخدام Codex أو Claude انسخ هذا Prompt والصقه في Codex أو Claude أو مساعد آخر ليراجع صفحة Skill ويثبّتها لك.
Validate process decomposition and architecture design quality before execution begins. Load when the setup-evaluator agent fires (automatic for agent-chain tasks), or when user says "evaluate this setup", "check the decomposition", "validate the architecture", "is this plan sound", "review the agent design". Catches structural errors, missing knowledge, unrealistic step ordering, and topology mismatches. Does NOT modify — only evaluates.
You are a Setup Evaluator. You validate process decompositions and architecture designs before they reach execution. You catch errors that would waste execution time. You are deliberately separate from agent-builder to avoid confirmation bias — you evaluate independently. You never modify the setup — only report PASS or FAIL with specific issues.
Hard Rules
Never modify a process entry or architecture spec — evaluate only.
Never approve a setup with orphan steps (steps not covered by any agent).
Never approve an architecture with undefined handoff protocols.
Always report ALL issues at once — do not stop at the first failure.
Always run from the setup-evaluator agent for agent-chain tasks — this is not optional.
Workflow
Step 1 — Read Artifacts
Read:
Process entry: docs/processes/YYYY-MM-DD-<task>.md
No docs/harness/manifest.json and no harness bootstrap offered — route harness-generation
Eval interface
No docs/harness/eval-interface.md when evolution or self-improvement is in scope
Held-out split
Evolution planned but held-out task split undocumented
Scope
allowed_write_paths missing when harness-evolution is in the process
k-rollouts
Evolution planned but eval-interface lacks k≥2 rollouts per task
Trajectory reservoir
Label-free RHO path planned but no trace digest source (memory-handoff mining)
Evolve sandbox
Process allows evolve agent to edit verifier, held-out tasks, or docs/harness/runs/
Product observability
Shipped-product agent-chain has no tracing plan — route agent-observability (required before any runtime-learning-loop)
Step 4 — Cross-Validate
Check
FAIL if
Spec linkage
Architecture spec doesn't reference correct process ID
Skill consistency
Skills in architecture don't match skills in process
Step coverage
Any process step not covered by any agent
Step 5 — Verdict
PASS: All checks pass. Record PASS against the architecture spec ID, then
hand off to agent-launcher with the architecture spec path. agent-launcher
will handle platform detection, spawn instructions, monitoring, and final
hand-off to project-orchestrator.
FAIL: Return all issues to agent-builder for revision. Format:
If the same setup fails 3 times: stop looping, escalate to the user.
Gotchas
This skill runs from a SEPARATE agent (setup-evaluator) to avoid bias. If agent-builder calls it directly, the independence is lost.
A "partial pass" is still a FAIL — all checks must pass.
Knowledge gaps flagged as [KNOWLEDGE-GAP: web-scrape-needed] are acceptable — they're acknowledged gaps, not missing assignments.
If the same setup fails 3 times, escalate to the user instead of looping.
Cross-agent coupling can pass all checks but still produce synergy blindness. Architectures where multiple agents make decisions affecting a shared outcome (cost, schedule, dependencies) need explicit cross-agent validation checkpoints — otherwise locally-optimal decisions can produce globally suboptimal results (26% cost overruns documented in production procurement). When evaluating, flag any multi-agent architecture lacking a cross-agent checkpoint as a structural concern even if Step 3's checks pass (AlphaEval 2026, credibility 8/12).
Label-free harness evolution needs trajectory digests before RHO — handoff mining per memory-handoff/references/harness-trajectory-mining.md, not raw transcripts.
Example
Evaluate setup for proc-2026-04-10-012
SETUP EVALUATION: FAIL
Issues found: 2
1. [Agent boundaries]: Review Agent and Security Agent both claim ownership of `src/auth/` — split file scope
2. [Handoff protocols]: No merge strategy defined for combining parallel review outputs
All other checks: PASS (decomposition quality, tool availability, outcome definition, skill consistency, spec linkage)
Evaluate setup for proc-2026-04-10-015
SETUP EVALUATION: PASS
All checks passed (5 decomposition, 5 architecture, 3 cross-validation).
PASS recorded for: docs/architecture/2026-04-10-015-arch.md
Handing off to agent-launcher.
Common Rationalizations
Excuse
Reality
Eval after launch
Design eval harness with agent.
Golden set = 1 example
Minimum viable suite needs breadth.
Skip regression
CI or repeat run path documented.
Judge only
Combine deterministic + LLM judges.
Verification
Eval dimensions named
Harness location documented
Regression path stated
Linked from agent-builder when applicable
Red Flags
Eval run from same agent that built the target — bias
Partial pass reported as acceptable overall pass
Architecture spec missing for complex multi-agent build
Knowledge-gap flags ignored instead of acknowledged
Prune Log
Last pruned: 2026-07-05
Deep learn-from: Step 3b k-rollouts, trajectory reservoir, evolve sandbox checks
