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/evidence/evidence_row.schema.json
View on GitHub{
"$schema": "https://json-schema.org/draft/2020-12/schema",
"$id": "https://github.com/Imbad0202/academic-research-skills/shared/contracts/evidence/evidence_row.schema.json",
"title": "ARS Shared Evidence Row v1",
"description": "Closed shared evidence-row carrier owned by #656. V1 persists one row per (Phase E claim, citation anchor) in Integrity Report phases.E_claims.evidence_rows[]. excerpt.state is provenance/availability only and MUST NOT be interpreted as a Phase E support verdict, severity, read attestation, or gate decision. Cross-field, exact-substring, word-budget, hash, strict-percent-decoding, and replay invariants are enforced by scripts/evidence_rows.py.",
"type": "object",
"additionalProperties": false,
"required": [
"schema_version",
"surface",
"row_id",
"claim",
"source",
"anchor",
"verdict",
"detail",
"excerpt",
"cache",
"content_handling",
"row_sha256"
],
"properties": {
"schema_version": {
"const": "evidence-row/1.0"
},
"surface": {
"const": "phase_e_claim_verification"
},
"row_id": {
"type": "string",
"pattern": "^EVR-[A-Za-z0-9._:-]+(?![\\s\\S])",
"maxLength": 200
},
"claim": {
"type": "object",
"additionalProperties": false,
"required": ["claim_id", "text", "paper_locator", "selection_tier"],
"properties": {
"claim_id": {
"type": "string",
"minLength": 1,
"maxLength": 200
},
"text": {
"type": "string",
"minLength": 1,
"maxLength": 2000
},
"paper_locator": {
"type": "string",
"minLength": 1,
"maxLength": 500
},
"selection_tier": {
"enum": ["HIGH-IMPACT", "RANDOM", "TOP-UP", "ALL"]
}
}
},
"source": {
"type": "object",
"additionalProperties": false,
"required": [
"ref_slug",
"display_label",
"source_content_sha256",
"source_content_utf8_bytes",
"source_artifact_sha256"
],
"properties": {
"ref_slug": {
"type": ["string", "null"],
"pattern": "^[A-Za-z][A-Za-z0-9_:-]*(?![\\s\\S])",
"maxLength": 300
},
"display_label": {
"type": ["string", "null"],
"minLength": 1,
"maxLength": 500
},
"source_content_sha256": {
"type": ["string", "null"],
"pattern": "^[0-9a-f]{64}(?![\\s\\S])",
"minLength": 64,
"maxLength": 64
},
"source_content_utf8_bytes": {
"type": ["integer", "null"],
"minimum": 0
},
"source_artifact_sha256": {
"type": ["string", "null"],
"pattern": "^[0-9a-f]{64}(?![\\s\\S])",
"minLength": 64,
"maxLength": 64,
"description": "Optional SHA-256 of raw artifact bytes. It is distinct from, and never substitutes for, source_content_sha256 over the exact session-held Unicode text used for matching. source_pointer is deliberately not persisted."
}
}
},
"anchor": {
"type": "object",
"additionalProperties": false,
"required": ["kind", "value_encoded", "value_decoded"],
"properties": {
"kind": {
"enum": ["quote", "page", "section", "paragraph", "none"]
},
"value_encoded": {
"type": "string",
"maxLength": 12000,
"description": "Writer-emitted marker payload. The runtime performs one strict RFC-3986 percent decode; '+' remains '+', malformed escapes and invalid UTF-8 are rejected, and rendering never decodes again."
},
"value_decoded": {
"type": "string",
"maxLength": 1000
}
}
},
"verdict": {
"enum": [
"VERIFIED",
"MINOR_DISTORTION",
"MAJOR_DISTORTION",
"UNVERIFIABLE",
"UNVERIFIABLE_ACCESS"
],
"description": "Existing Phase E claim verdict, carried without modification. There is deliberately no conditional mapping from verdict to excerpt.state."
},
"detail": {
"type": "string",
"maxLength": 2000
},
"excerpt": {
"type": "object",
"additionalProperties": false,
"required": [
"state",
"text",
"excerpt_sha256",
"source_span_utf8",
"captured_at"
],
"properties": {
"state": {
"enum": [
"verified_exact_match",
"agent_extracted",
"unconfirmed_anchor",
"not_checked",
"source_missing",
"access_failed",
"retrieval_failed",
"anchorless"
],
"description": "The single canonical evidence provenance/availability vocabulary. It never expresses whether the claim is supported."
},
"text": {
"type": ["string", "null"],
"minLength": 1,
"maxLength": 1000,
"description": "At most 25 words by Python whitespace split, additionally bounded to 1000 Unicode code points. Non-null only for verified_exact_match or agent_extracted."
},
"excerpt_sha256": {
"type": ["string", "null"],
"pattern": "^[0-9a-f]{64}(?![\\s\\S])",
"minLength": 64,
"maxLength": 64
},
"source_span_utf8": {
"oneOf": [
{
"type": "object",
"additionalProperties": false,
"required": ["start", "end"],
"properties": {
"start": {"type": "integer", "minimum": 0},
"end": {"type": "integer", "minimum": 1}
}
},
{"type": "null"}
],
"description": "Zero-based, end-exclusive byte offsets in source_content_sha256's exact UTF-8 byte stream."
},
"captured_at": {
"type": ["string", "null"],
"format": "date-time",
"pattern": "^[0-9]{4}-[0-9]{2}-[0-9]{2}[Tt](?:[01][0-9]|2[0-3]):[0-5][0-9]:[0-5][0-9](?:\\.[0-9]+)?(?:[Zz]|[+-](?:[01][0-9]|2[0-3]):[0-5][0-9])(?![\\s\\S])",
"description": "Original extraction timestamp. Cache replay preserves it instead of refreshing it."
}
}
},
"cache": {
"type": "object",
"additionalProperties": false,
"required": ["status", "key_sha256"],
"properties": {
"status": {
"enum": ["not_used", "miss", "hit"]
},
"key_sha256": {
"type": ["string", "null"],
"pattern": "^[0-9a-f]{64}(?![\\s\\S])",
"minLength": 64,
"maxLength": 64
}
}
},
"content_handling": {
"type": "object",
"additionalProperties": false,
"required": ["contains_external_text", "sharing_scope", "rights_basis"],
"properties": {
"contains_external_text": {
"type": "boolean",
"description": "True when the row persists an excerpt or a decoded quote anchor. This is derived, not a rights judgment."
},
"sharing_scope": {
"enum": ["session_only", "user_confirmed_shareable"]
},
"rights_basis": {
"enum": ["not_assessed", "user_declared_authorized"]
}
},
"description": "External-text rows default to session_only/not_assessed. The 25-word ceiling is data minimization, not a licence or fair-use determination."
},
"row_sha256": {
"type": "string",
"pattern": "^[0-9a-f]{64}(?![\\s\\S])",
"minLength": 64,
"maxLength": 64,
"description": "Unkeyed SHA-256 integrity checksum of the runtime's canonical JSON serialization of the complete row excluding only row_sha256 itself. It is not a signature or independent provenance proof; source-bound rows require explicit source replay before rendering."
}
},
"allOf": [
{
"if": {
"properties": {
"content_handling": {
"properties": {"sharing_scope": {"const": "user_confirmed_shareable"}},
"required": ["sharing_scope"]
}
}
},
"then": {
"properties": {
"content_handling": {
"properties": {"rights_basis": {"const": "user_declared_authorized"}}
}
}
}
},
{
"if": {
"properties": {
"content_handling": {
"properties": {"rights_basis": {"const": "user_declared_authorized"}},
"required": ["rights_basis"]
}
}
},
"then": {
"properties": {
"content_handling": {
"properties": {"sharing_scope": {"const": "user_confirmed_shareable"}}
}
}
}
},
{
"if": {
"properties": {
"anchor": {
"properties": {"kind": {"const": "none"}},
"required": ["kind"]
}
}
},
"then": {
"properties": {
"anchor": {
"properties": {
"value_encoded": {"const": ""},
"value_decoded": {"const": ""}
}
},
"excerpt": {
"properties": {"state": {"const": "anchorless"}}
}
}
}
},
{
"if": {
"properties": {
"anchor": {
"properties": {"kind": {"const": "paragraph"}},
"required": ["kind"]
}
}
},
"then": {
"properties": {
"excerpt": {
"properties": {"state": {"const": "not_checked"}}
}
}
}
},
{
"if": {
"properties": {
"excerpt": {
"properties": {
"state": {"enum": ["verified_exact_match", "agent_extracted"]}
},
"required": ["state"]
}
}
},
"then": {
"properties": {
"source": {
"properties": {
"ref_slug": {"type": "string"},
"source_content_sha256": {"type": "string"},
"source_content_utf8_bytes": {"type": "integer"}
}
},
"excerpt": {
"properties": {
"text": {"type": "string", "minLength": 1},
"excerpt_sha256": {"type": "string"},
"source_span_utf8": {"type": "object"},
"captured_at": {"type": "string", "format": "date-time"}
}
},
"cache": {
"properties": {
"status": {"enum": ["miss", "hit"]},
"key_sha256": {"type": "string"}
}
}
}
}
},
{
"if": {
"properties": {
"excerpt": {
"properties": {"state": {"const": "verified_exact_match"}},
"required": ["state"]
}
}
},
"then": {
"properties": {
"anchor": {"properties": {"kind": {"const": "quote"}}}
}
}
},
{
"if": {
"properties": {
"excerpt": {
"properties": {"state": {"const": "agent_extracted"}},
"required": ["state"]
}
}
},
"then": {
"properties": {
"anchor": {"properties": {"kind": {"enum": ["page", "section"]}}}
}
}
},
{
"if": {
"properties": {
"excerpt": {
"properties": {"state": {"const": "unconfirmed_anchor"}},
"required": ["state"]
}
}
},
"then": {
"properties": {
"anchor": {"properties": {"kind": {"const": "quote"}}},
"source": {
"properties": {
"ref_slug": {"type": "string"},
"source_content_sha256": {"type": "string"},
"source_content_utf8_bytes": {"type": "integer"}
}
},
"excerpt": {
"properties": {
"text": {"type": "null"},
"excerpt_sha256": {"type": "null"},
"source_span_utf8": {"type": "null"},
"captured_at": {"type": "null"}
}
},
"cache": {
"properties": {
"status": {"enum": ["miss", "hit"]},
"key_sha256": {"type": "string"}
}
}
}
}
},
{
"if": {
"properties": {
"excerpt": {
"properties": {
"state": {
"enum": [
"not_checked",
"source_missing",
"access_failed",
"retrieval_failed",
"anchorless"
]
}
},
"required": ["state"]
}
}
},
"then": {
"properties": {
"source": {
"properties": {
"source_content_sha256": {"type": "null"},
"source_content_utf8_bytes": {"type": "null"}
}
},
"excerpt": {
"properties": {
"text": {"type": "null"},
"excerpt_sha256": {"type": "null"},
"source_span_utf8": {"type": "null"},
"captured_at": {"type": "null"}
}
},
"cache": {
"properties": {
"status": {"const": "not_used"},
"key_sha256": {"type": "null"}
}
}
}
}
},
{
"if": {
"properties": {
"anchor": {
"properties": {"kind": {"enum": ["page", "section"]}},
"required": ["kind"]
}
}
},
"then": {
"properties": {
"excerpt": {
"properties": {
"state": {
"enum": [
"agent_extracted",
"not_checked",
"source_missing",
"access_failed",
"retrieval_failed"
]
}
}
}
}
}
},
{
"if": {
"properties": {
"excerpt": {
"properties": {"state": {"const": "anchorless"}},
"required": ["state"]
}
}
},
"then": {
"properties": {
"anchor": {
"properties": {
"kind": {"const": "none"},
"value_encoded": {"const": ""},
"value_decoded": {"const": ""}
}
}
}
}
}
]
}
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 |