zrwangems/dsh-zrwang-plugin / ars/skills/academic-pipeline
academic-pipeline
Flagged学术研究总编排:研究→写作→诚信检查→评审→返修→复审→终检→定稿的十阶段流水线,每阶段停在用户确认点,含材料护照与 7 类 AI 研究失败模式检查。Triggers: academic pipeline, research to paper, full paper workflow, end-to-end paper | 完整论文流程, 从研究到论文, 端到端写论文, 论文流水线。
_shared/contracts/passport/claim_drift.schema.json
View on GitHub{
"$schema": "https://json-schema.org/draft/2020-12/schema",
"$id": "https://github.com/Imbad0202/academic-research-skills/shared/contracts/passport/claim_drift.schema.json",
"title": "Material Passport Claim Drift Entry",
"description": "One entry in Material Passport Schema 9 claim_drifts[] aggregate. Per claim-intent-drift finding emitted by claim_ref_alignment_audit_agent §4 step 5 manifest set-diff. Rides in its own aggregate because claim-intent drift is detected by manifest set-diff (intended ≠ emitted), NOT by judge invocation — embedding drift findings as a defect_stage of claim_audit_result would force judgment=UNSUPPORTED to be recorded without the judge ever evaluating that claim, contaminating both the schema semantics and the calibration FNR/FPR metrics. Always LOW-WARN advisory at the finalizer (per D4-a — gate-refuse is reserved for explicit author negative constraints, not drift signals). Cross-field invariants D-INV-1..D-INV-4 are lint-enforced in scripts/check_claim_audit_consistency.py. See docs/design/2026-05-15-issue-103-claim-alignment-audit-spec.md §3.4.",
"type": "object",
"additionalProperties": false,
"required": [
"finding_id",
"drift_kind",
"claim_text",
"detected_at",
"rule_version"
],
"properties": {
"finding_id": {
"type": "string",
"pattern": "^CD-[0-9]{3,}$",
"description": "Per-passport unique finding identifier for claim_drifts[]. D-INV-1 enforces uniqueness."
},
"drift_kind": {
"enum": ["EMITTED_NOT_INTENDED", "INTENDED_NOT_EMITTED"],
"description": "EMITTED_NOT_INTENDED: an emitted claim that was not in the manifest (the drafted prose introduced a claim the writer did not pre-commit to). INTENDED_NOT_EMITTED: a manifest claim that did not appear in the emitted prose (the writer dropped a pre-committed claim during drafting). Both are advisory — gate-refuse is reserved for negative_constraint_violation."
},
"claim_text": {
"type": "string",
"minLength": 1,
"maxLength": 2000,
"description": "For EMITTED_NOT_INTENDED, the emitted sentence; for INTENDED_NOT_EMITTED, the manifest claim_text."
},
"manifest_claim_id": {
"type": ["string", "null"],
"pattern": "^C-[0-9]{3,}$",
"description": "For INTENDED_NOT_EMITTED, the dropped manifest claim_id (REQUIRED, conditional on drift_kind — enforced by D-INV-2). For EMITTED_NOT_INTENDED, null (drifted claim has no manifest_claim_id since it was never in the manifest). When present, MUST be paired with scoped_manifest_id."
},
"scoped_manifest_id": {
"type": ["string", "null"],
"pattern": "^M-[0-9]{4}-[0-9]{2}-[0-9]{2}T[0-9]{2}:[0-9]{2}:[0-9]{2}Z-[0-9a-f]{4}$",
"description": "Points to the claim_intent_manifest.manifest_id owning the referenced manifest_claim_id. Required when drift_kind=INTENDED_NOT_EMITTED (paired with manifest_claim_id per D-INV-2 cross-array integrity, disambiguating C-001 collision across manifests). Null when drift_kind=EMITTED_NOT_INTENDED (the drifted claim has no manifest origin)."
},
"section_path": {
"type": ["string", "null"],
"minLength": 1,
"description": "For EMITTED_NOT_INTENDED, hierarchical path from document root to the section containing the drifted sentence (mirrors uncited_assertion.section_path). Null for INTENDED_NOT_EMITTED (dropped claim has no draft location)."
},
"detected_at": { "type": "string", "format": "date-time" },
"rule_version": {
"const": "D4-a-v1",
"description": "Frozen literal for the v3.8.0 release. D-INV-3 enforces equality. Future rule revisions bump the constant and require re-lint."
},
"upstream_owner_agent": {
"enum": ["synthesis_agent", "draft_writer_agent", "report_compiler_agent", null]
}
}
}
Related skills
| Skill | Stars | Updated | Works with | Status | Install |
|---|---|---|---|---|---|
| writing-skillsobra/superpowersUse when creating new skills, editing existing skills, or verifying skills work before deployment | Stars: 322k | Updated | Claude CodeCodex | Verified | Install |
| xlsxanthropics/skillsUse this skill any time a spreadsheet file is the primary input or output. This means any task where the user wants to: open, read, edit, or fix an existing .xlsx, .xlsm, .xltx, .csv, or .tsv file (e.g., adding columns, computing formulas, formatting, charting, cleaning messy data); create a new spreadsheet from scratch or from other data sources; or convert between tabular file formats. Trigger especially when the user references a spreadsheet file by name or path — even casually (like "the xlsx in my downloads") — and wants something done to it or produced from it. Also trigger for cleaning or restructuring messy tabular data files (malformed rows, misplaced headers, junk data) into proper spreadsheets. The deliverable must be a spreadsheet file. Do NOT trigger when the primary deliverable is a Word document, HTML report, standalone Python script, database pipeline, or Google Sheets API integration, even if tabular data is involved. | Stars: 198k | Updated | Claude CodeCodex | Verified | Install |
| shipping-and-launchaddyosmani/agent-skillsPrepares production launches. Use when preparing to deploy to production, or when asking what needs to be in place before shipping. Use when you need a pre-launch checklist, when setting up monitoring, when planning a staged rollout, or when you need a rollback strategy. | Stars: 119k | Updated | Claude CodeCodex | Verified | Install |
| childhoodthedotmack/claude-memBabysit a shipped release after merge, publication, or deployment. Verify published packages, production behavior, installed plugins, and late CI or user reports until the observation window is clean; use babysit for PR readiness before release. | Stars: 118k | Updated | Claude CodeCodex | Verified | Install |