result-to-claim

name	result-to-claim
description	Use when experiments complete to judge what claims the results support, what they don't, and what evidence is still missing. local coding-assistant integration evaluates results against intended claims and routes to next action (pivot, supplement, or confirm). Use after experiments finish — before writing the paper or running ablations.
argument-hint	["experiment-description-or-wandb-run"]

Web-side execution adapter

• This skill is workflow guidance for the ChatGPT web-side connector.
• Loading this SKILL.md is only the setup step; it does not mean the task is complete.
• After loading, continue to execute the workflow, constraints, and output format below before answering.
• Mentions of local automation, local file operations, local command execution, or external integrations are descriptive only. Use capabilities available in the current ChatGPT session, or ask the user for needed files/links.
• For literature search, current facts, factual verification, source tracing, numeric values, material properties, legal/medical/financial/current information, or any evidence-heavy claim: use available search/browsing tools first and cite verifiable sources. Do not answer such tasks only from memory.
• Preserve the original workflow and scope unless the user explicitly asks for changes.

Result-to-Claim Gate

Experiments produce numbers; this gate decides what those numbers mean. Collect results from available sources, get a local coding assistant judgment, then auto-route based on the verdict.

Context: $ARGUMENTS

When to Use

• After a set of experiments completes (main results, not just sanity checks)
• Before committing to claims in a paper or review response
• When results are ambiguous and you need an objective second opinion

Workflow

Step 1: Collect Results

Gather experiment data from whatever sources are available in the project:

1. W&B (preferred): wandb.Api().run("<entity>/<project>/<run_id>").history() — metrics, training curves, comparisons
2. EXPERIMENT_LOG.md: full results table with baselines and verdicts
3. EXPERIMENT_TRACKER.md: check which experiments are DONE vs still running
4. Log files: ssh server "tail -100 /path/to/training.log" if no other source
5. docs/research_contract.md: intended claims and experiment design

Assemble the key information:

• What experiments were run (method, dataset, config)
• Main metrics and baseline comparisons (deltas)
• The intended claim these experiments were designed to test
• Any known confounds or caveats

Step 2: local coding assistant Judgment

Send the collected results to local coding assistant for objective evaluation:

local coding assistant integration:
  config: {"model_reasoning_effort": "xhigh"}
  prompt: |
    RESULT-TO-CLAIM EVALUATION

    I need you to judge whether experimental results support the intended claim.

    Intended claim: [the claim these experiments test]

    Experiments run:
    [list experiments with method, dataset, metrics]

    Results:
    [paste key numbers, comparison deltas, significance]

    Baselines:
    [baseline numbers and sources — reproduced or from paper]

    Known caveats:
    [any confounding factors, limited datasets, missing comparisons]

    Please evaluate:
    1. claim_supported: yes | partial | no
    2. what_results_support: what the data actually shows
    3. what_results_dont_support: where the data falls short of the claim
    4. missing_evidence: specific evidence gaps
    5. suggested_claim_revision: if the claim should be strengthened, weakened, or reframed
    6. next_experiments_needed: specific experiments to fill gaps (if any)
    7. confidence: high | medium | low

    Be honest. Do not inflate claims beyond what the data supports.
    A single positive result on one dataset does not support a general claim.

Step 3: Parse and Normalize

Extract structured fields from local coding assistant response:

- claim_supported: yes | partial | no
- what_results_support: "..."
- what_results_dont_support: "..."
- missing_evidence: "..."
- suggested_claim_revision: "..."
- next_experiments_needed: "..."
- confidence: high | medium | low

Step 3.5: Check Experiment Integrity (if audit exists)

Skip this step if EXPERIMENT_AUDIT.json does not exist.

if EXPERIMENT_AUDIT.json exists:
    read integrity_status from file
    attach to verdict output:
        integrity_status: pass | warn | fail

    if integrity_status == "fail":
        append to verdict: "[INTEGRITY CONCERN] — audit found issues, see EXPERIMENT_AUDIT.md"
        downgrade confidence to "low" regardless of local coding assistant judgment

    if integrity_status == "warn":
        append to verdict: "[INTEGRITY: WARN] — audit flagged potential issues"
else:
    integrity_status = "unavailable"
    verdict is labeled "provisional — no integrity audit run"
    (this does NOT block anything — pipeline continues normally)

See shared-references/experiment-integrity.md for the full integrity protocol.

Step 4: Route Based on Verdict

no — Claim not supported

1. Record postmortem in findings.md (Research Findings section):
   • What was tested, what failed, hypotheses for why
   • Constraints for future attempts (what NOT to try again)
2. Update CLAUDE.md Pipeline Status
3. Decide whether to pivot to next idea from IDEA_CANDIDATES.md or try an alternative approach

partial — Claim partially supported

1. Update the working claim to reflect what IS supported
2. Record the gap in findings.md
3. Design and run supplementary experiments to fill evidence gaps
4. Re-run result-to-claim after supplementary experiments complete
5. Multiple rounds of partial on the same claim → record analysis in findings.md, consider whether to narrow the claim scope or switch ideas

yes — Claim supported

1. Record confirmed claim in project notes
2. If ablation studies are incomplete → trigger /ablation-planner
3. If all evidence is in → ready for paper writing

Step 5: Update Research Wiki (if active)

Skip this step entirely if research-wiki/ does not exist.

if research-wiki/ exists:
    # 1. Create experiment page
    Create research-wiki/experiments/<exp_id>.md with:
      - node_id: exp:<id>
      - idea_id: idea:<active_idea>
      - date, hardware, duration, metrics
      - verdict, confidence, reasoning summary

    # 2. Update claim status
    for each claim resolved by this verdict:
        if verdict == "yes":
            Update claim page: status → supported
            python3 tools/research_wiki.py add_edge research-wiki/ --from "exp:<id>" --to "claim:<cid>" --type supports --evidence "<metric>"
        elif verdict == "partial":
            Update claim page: status → partial
            python3 tools/research_wiki.py add_edge research-wiki/ --from "exp:<id>" --to "claim:<cid>" --type supports --evidence "partial"
        else:
            Update claim page: status → invalidated
            python3 tools/research_wiki.py add_edge research-wiki/ --from "exp:<id>" --to "claim:<cid>" --type invalidates --evidence "<why>"

    # 3. Update idea outcome
    Update research-wiki/ideas/<idea_id>.md:
      - outcome: positive | mixed | negative
      - If negative: fill "Failure / Risk Notes" and "Lessons Learned"
      - If positive: fill "Actual Outcome" and "Reusable Components"

    # 4. Rebuild + log
    python3 tools/research_wiki.py rebuild_query_pack research-wiki/
    python3 tools/research_wiki.py log research-wiki/ "result-to-claim: exp:<id> verdict=<verdict> for idea:<idea_id>"

    # 5. Re-ideation suggestion
    Count failed/partial ideas since last /idea-creator run.
    If >= 3: print "💡 3+ ideas tested since last ideation. Consider re-running /idea-creator — the wiki now knows what doesn't work."

Rules

• local coding assistant is the judge, not CC. CC collects evidence and routes; local coding assistant evaluates. This prevents post-hoc rationalization.
• Do not inflate claims beyond what the data supports. If local coding assistant says "partial", do not round up to "yes".
• A single positive result on one dataset does not support a general claim. Be honest about scope.
• If confidence is low, treat the judgment as inconclusive and add experiments rather than committing to a claim.
• If local coding-assistant integration is unavailable (call fails), CC makes its own judgment and marks it [pending local coding assistant review] — do not block the pipeline.
• Always record the verdict and reasoning in findings.md, regardless of outcome.