Eval Harness

March 6, 2026 · View on GitHub

🧭 Quick Return to Map

You are in a sub-page of Eval.
To reorient, go back here:

Think of this page as a desk within a ward.
If you need the full triage and all prescriptions, return to the Emergency Room lobby.

Evaluation disclaimer (eval harness)
This page sketches a harness for running structured evaluations on AI pipelines.
Any metrics or labels that pass through such a harness remain heuristic outputs of models, scripts and annotators.
They do not become scientific proof just because they flow through this structure.
Use the harness to compare variants inside a controlled scenario, and avoid presenting those numbers as universal claims about model quality beyond that scenario.

A minimal yet strict harness to run repeatable evaluations for RAG and agent pipelines. It fixes the two usual failures. First, non-reproducible runs. Second, noisy metrics that cannot explain drift. Everything here maps to WFGY pages with measurable targets.

Open these first

Acceptance targets for this harness

  • ΔS(question, retrieved) ≤ 0.45 on the gold set
  • Coverage of the target section ≥ 0.70
  • λ remains convergent across 3 paraphrases and 2 seeds
  • Re-runs with identical seed produce metrics drift ≤ 0.5 percentage point

Folder layout and contracts

eval/
  datasets/
    gold/
      qa.jsonl            # minimal gold set
      citations.jsonl     # expected snippet anchors
    probes/
      paraphrases.jsonl   # 3 paraphrases per item
  runs/
    2025-08-29_seed42/
      config.yaml
      metrics.csv
      traces.jsonl
  config/
    harness.yaml          # store, retriever, reranker, seeds, k

Input schema

datasets/gold/qa.jsonl one JSON per line.

{
  "id": "Q_0001",
  "question": "How is vector contamination detected in FAISS indexes",
  "answer_ref": "PM:vectorstore-metrics-and-faiss-pitfalls#detect-contamination",
  "expected_doc": "ProblemMap/vectorstore-metrics-and-faiss-pitfalls.md",
  "section_id": "detect-contamination"
}

datasets/gold/citations.jsonl

{
  "id": "Q_0001",
  "snippet_id": "S_18823",
  "section_id": "detect-contamination",
  "source_url": "https://github.com/onestardao/WFGY/blob/main/ProblemMap/vectorstore-metrics-and-faiss-pitfalls.md",
  "offsets": [1380, 1540],
  "tokens": [310, 352]
}

Contract rules come from Retrieval Traceability and Data Contracts.

Repro knobs

  • seed: integer. Set for the retriever, reranker, and LLM sampler if available.
  • k: top k per retriever. Test 5, 10, 20.
  • λ_observe: record λ state for retrieve, assemble, reason. See lambda_observe.md.
  • ΔS probe: compute ΔS(question, retrieved) and ΔS(retrieved, expected anchor). See deltaS_thresholds.md.

Execution flow

  1. Warm up fence. Verify index hash, vector ready, secrets. If not ready, stop. Open: Bootstrap Ordering.

  2. Retrieval step. Run with fixed metric and analyzer. Save raw hits with snippet fields from the contract page.

  3. ΔS and λ probes. Log both per item. If ΔS ≥ 0.60 flag as structural risk.

  4. Reasoning step. LLM reads TXT OS and uses the cite then explain schema. Refuse answers without citations.

  5. Metrics. Compute precision, recall, citation hit, coverage. See eval_rag_precision_recall.md and Retrieval Playbook.

  6. Trace sink. Write traces.jsonl with id, seed, k, ΔS, λ_state, snippet_id, section_id, INDEX_HASH.

  7. Gate. If coverage < 0.70 or ΔS > 0.45 fail the run. See regression_gate.md.

Sixty second quick start

  1. Place a ten item gold set into datasets/gold/qa.jsonl and citations.jsonl.
  2. Copy config/harness.yaml from a previous good run. Set seed: 42, k: 10.
  3. Run your script to produce runs/<date>_seed42/metrics.csv and traces.jsonl.
  4. Verify the acceptance targets above. If any gate fails jump to the right fix below.

Common failures and the exact fix

CI gates and artifacts

  • Block merge if any of these is true

    1. ΔS median > 0.45 on gold
    2. Coverage < 0.70
    3. λ flips on 2 of 3 paraphrases
    4. Metrics drift from last green run > 0.5 percentage point

  • Store artifacts metrics.csv, traces.jsonl, harness.yaml, INDEX_HASH, MODEL_HASH.

Copy paste prompts for the reasoning step

You have TXTOS and the WFGY Problem Map loaded.

Question: "{question}"
Retrieved snippets: [{snippet_id, section_id, source_url, offsets, tokens}]

Do:
1) Cite then explain. If citation is missing or mismatched, fail fast and return the minimal structural fix.
2) If ΔS(question, retrieved) ≥ 0.60 propose the smallest repair. Use retrieval-playbook, retrieval-traceability, data-contracts, rerankers.
3) Return JSON:
   {"citations":[...], "answer":"...", "λ_state":"→|←|<>|×", "ΔS":0.xx, "next_fix":"..."}
Keep it short and auditable.

🔗 Quick-Start Downloads (60 sec)

ToolLink3-Step Setup
WFGY 1.0 PDFEngine Paper1️⃣ Download · 2️⃣ Upload to your LLM · 3️⃣ Ask “Answer using WFGY + <your question>”
TXT OS (plain-text OS)TXTOS.txt1️⃣ Download · 2️⃣ Paste into any LLM chat · 3️⃣ Type “hello world” — OS boots instantly

Explore More

LayerPageWhat it’s for
⭐ ProofWFGY Recognition MapExternal citations, integrations, and ecosystem proof
⚙️ EngineWFGY 1.0Original PDF tension engine and early logic sketch (legacy reference)
⚙️ EngineWFGY 2.0Production tension kernel for RAG and agent systems
⚙️ EngineWFGY 3.0TXT based Singularity tension engine (131 S class set)
🗺️ MapProblem Map 1.0Flagship 16 problem RAG failure taxonomy and fix map
🗺️ MapProblem Map 2.0Global Debug Card for RAG and agent pipeline diagnosis
🗺️ MapProblem Map 3.0Global AI troubleshooting atlas and failure pattern map
🧰 AppTXT OS.txt semantic OS with fast bootstrap
🧰 AppBlah Blah BlahAbstract and paradox Q&A built on TXT OS
🧰 AppBlur Blur BlurText to image generation with semantic control
🏡 OnboardingStarter VillageGuided entry point for new users

If this repository helped, starring it improves discovery so more builders can find the docs and tools.
GitHub Repo stars