{"id":18416,"plugin_id":"plugins_6a893288a1008191857f5437d78ab047","kind":"skill","collection_source":null,"comparison_source":null,"observed_at":"2026-09-30T23:14:48.678Z","digest":"5e1d086d46d558dc79d5c7903562f0c2f8a803f182ce6bf89b72deda6281fb30","against":null,"payload":{"description":"Use before finalizing high-impact marketing recommendations or whenever multiple analyses disagree, evidence is weak, causal language may be overstated, or a live campaign change is being considered.","included_files":[{"relative_path":"agents/openai.yaml","size_in_bytes":300}],"name":"decision-quality-gate","skill_md_contents":"---\nname: decision-quality-gate\ndescription: Use before finalizing high-impact marketing recommendations or whenever multiple analyses disagree, evidence is weak, causal language may be overstated, or a live campaign change is being considered.\n---\n\n# Decision Quality Gate\n\nReview the recommendation, not just the prose.\n\n## Gate checks\n\n### Evidence integrity\n\n- Are stated observations present in the supplied or retrieved data?\n- Are calculations reproducible?\n- Are time windows and metric definitions compatible?\n- Are missing data and reporting lag acknowledged?\n\n### Causal discipline\n\n- Does the recommendation claim causality from correlation?\n- Could attribution overlap or confounding reverse the conclusion?\n- Is a modeled forecast being presented as observed fact?\n\n### Decision economics\n\n- Does the action align with the business objective rather than a proxy metric?\n- Are margin, volume, capacity, and opportunity cost relevant?\n- Is the proposed action material enough to matter but bounded enough to learn safely?\n\n### Risk and reversibility\n\nClassify the proposal:\n\n- low-risk / reversible\n- moderate-risk / reversible with monitoring\n- high-impact / requires stronger evidence or staged rollout\n- blocked / evidence or permissions insufficient\n\n### Execution boundary\n\nA plan is not an executed change. If the host provides a compatible write-capable app and the user explicitly authorizes the mutation, restate the exact change set before execution. Otherwise return a ready-to-execute plan only.\n\n## Scorecard\n\nScore each dimension `pass`, `warn`, or `fail`:\n\n- evidence\n- measurement\n- causal claim\n- expected economics\n- risk\n- reversibility\n- authorization\n\nAny `fail` on evidence, authorization, or a material causal claim blocks live execution.\n\n## Output contract\n\nReturn gate result, warnings, blocked claims/actions, revised recommendation, and the minimum additional evidence needed to clear the gate.\n"},"changes":[],"summary":"First saved snapshot. No earlier version is available for comparison.","summary_kind":"deterministic","summary_metadata":{}}