Deep Research — Universal Academic Research Agent Team

Universal deep research tool — a domain-agnostic 13-agent team for rigorous academic research on any topic.

v2.4 adds writing quality improvements to the report compiler:

• Style Profile consumption (optional) — If a Style Profile is available from ars-academic-paper intake, the report compiler applies it as a soft guide for the Executive Summary and Synthesis sections. Discipline conventions and report objectivity take priority.
• Writing Quality Check — The report compiler runs a writing quality checklist before finalizing: flags AI-typical overused terms, checks sentence/paragraph length variation, removes throat-clearing openers. See ars-academic-paper/references/guides/writing_quality_check.md.

Routing discipline: see references/shared/intent_clarification_protocol.md for cross-skill routing rules. This skill assumes routing has already settled — ambiguous cross-phase materials should have been clarified upstream.

Runtime Environment Self-Check (run first, non-blocking)

When this skill is loaded, the orchestrating session first checks the host runtime's capabilities. Never modify runtime configuration; never abort the task on a failed check — degrade gracefully instead.

Sub-Agent Dispatch Rules

1. When dispatching ANY role agent as a sub-session, the FIRST sentence of the task instruction MUST be: "Before executing, fully read the role rule file ./references/<role_file>.md and load all its constraints. Do not skip the role file; do not work from the task summary alone."
2. Sub-sessions start with isolated context — pass only the materials required for the task; do not assume the sub-agent sees the main session's history.
3. When a sub-agent returns, the orchestrator collects its output, closes the sub-session, then proceeds to the next step.
4. If the runtime has no sub-agent dispatch, execute each role inline (role-play) in the current session — the role file MUST still be read and applied with equal strictness.

Universal Academic Integrity Rules (binding on ALL agents)

1. Vision/OCR mandatory reading — scanned documents, figure tables, logic diagrams, and complex formulas MUST be parsed with vision/OCR tooling; never guess image content from surrounding text.
2. Evidence binding — every academic claim must carry a verifiable source; any unsupported conjecture MUST be labeled [SPECULATION — UNVERIFIED] / 【推测·待验证】.
3. Zero hallucination — never fabricate DOIs, references, authors, experimental data, or research results.
4. Logic conservation — never silently alter the agreed research scope, direction, or assumptions mid-project.
5. Computation recheck — after any mathematical derivation or data computation, remind the user to manually verify the result.

Quick Start

Minimal command:

Research the impact of AI on higher education quality assurance

Socratic mode:

Guide my research on the impact of declining birth rates on private universities
引導我的研究：少子化對私立大學的影響
幫我釐清我的研究方向，我對高教品保有興趣但還不太確定

Execution:

1. Scoping — Research question + methodology blueprint
2. Investigation — Systematic literature search + source verification
3. Analysis — Cross-source synthesis + bias check
4. Composition — Full APA 7.0 report
5. Review — Editorial + ethics + vulnerability scan
6. Revision — Final polished report

Trigger Conditions

Trigger Keywords

English: research, deep research, literature review, systematic review, meta-analysis, PRISMA, evidence synthesis, fact-check, methodology, APA report, academic analysis, policy analysis, WHY HOW WHAT papers, 3W literature scan, guide my research, help me think through, monitor this topic, set up alerts

繁體中文: 研究, 深度研究, 文獻回顧, 文獻探討, 系統性回顧, 後設分析, 證據綜整, 事實查核, 三段式文獻掃描, WHY HOW WHAT 論文比較, 研究方法, 學術分析, 政策分析, 引導我的研究, 幫我釐清, 監測這個主題, 設定追蹤

한국어: 심층 연구, 문헌 조사, 문헌 고찰, 체계적 문헌고찰, 메타분석, 근거 종합, 사실 확인, 팩트체크, 연구 방법 설계, 학술 분석, 연구 방향을 잡아줘, 연구 주제 정하는 것을 도와줘, 무엇을 연구할지 모르겠어, 이 주제 계속 모니터링해줘

Socratic Mode Activation

Activate socratic mode when the user's intent matches any of the following patterns, regardless of language. Detect meaning, not exact keywords.

Intent signals (any one is sufficient):

1. User has no clear research question and wants guided thinking
2. User asks to be "led", "guided", or "mentored" through research
3. User expresses uncertainty about what to research or where to start
4. User wants to brainstorm, explore, or clarify a research direction
5. User describes a vague interest without a specific, answerable question

Default rule: When intent is ambiguous between socratic and full, prefer socratic — it is safer to guide first than to produce an unwanted report. The user can always switch to full later.

Example triggers (illustrative, not exhaustive): "guide my research", "help me think through", 「引導我的研究」「幫我釐清」, or equivalent in any language

Does NOT Trigger

Quick Mode Selection Guide

Spectrum (v3.2): fidelity = template-heavy, predictable output; balanced = default; originality = exploratory, template-light. See references/shared/mode_spectrum.md for the full cross-skill spectrum table.

Not sure? Start with socratic — it will help you figure out what you need. 不確定？先用 socratic 模式——它會幫你釐清你需要什麼。

Agent Team (13 Agents)

Mode Selection Guide

See references/guides/mode_selection_guide.md for the detailed guide.

User Input
    |
    +-- Already have a clear research question?
    |   +-- Yes --> Need PRISMA-compliant systematic review / meta-analysis?
    |   |           +-- Yes --> systematic-review mode
    |   |           +-- No --> Need a full report?
    |   |                      +-- Yes --> full mode
    |   |                      +-- No --> Only need literature?
    |   |                                 +-- Yes --> Need rapid paper comparison?
    |   |                                            +-- Yes --> three-way-scan mode
    |   |                                            +-- No --> lit-review mode
    |   |                                 +-- No --> quick mode
    |   +-- No --> Want to be guided through thinking?
    |              +-- Yes --> socratic mode
    |              +-- No --> full mode (Phase 1 will be interactive)
    |
    +-- Already have text to review? --> review mode
    +-- Only need fact-checking? --> fact-check mode

Orchestration Workflow (6 Phases)

User: "Research [topic]"
     |
=== Phase 1: SCOPING (Interactive) ===
     |
     |-> [research_question_agent] -> RQ Brief
     |   - FINER criteria scoring (Feasible, Interesting, Novel, Ethical, Relevant)
     |   - Scope boundaries (in-scope / out-of-scope)
     |   - 2-3 sub-questions
     |
     |-> [research_architect_agent] -> Methodology Blueprint
     |   - Research paradigm (positivist / interpretivist / pragmatist)
     |   - Method selection (qualitative / quantitative / mixed)
     |   - Data strategy (primary / secondary / both)
     |   - Analytical framework
     |   - Validity & reliability criteria
     |
     +-> [devils_advocate_agent] -- CHECKPOINT 1
         - RQ clarity and answerable?
         - Method appropriate for question?
         - Scope too broad or too narrow?
         - Verdict: PASS / REVISE (with specific feedback)
     |
     ** User confirmation before Phase 2 **
     |
=== Phase 2: INVESTIGATION ===
     |
     |-> [bibliography_agent] -> Source Corpus + Annotated Bibliography
     |   - Systematic search strategy (databases, keywords, Boolean)
     |   - Inclusion/exclusion criteria
     |   - PRISMA-style flow (if applicable)
     |   - Annotated bibliography (APA 7.0)
     |
     +-> [source_verification_agent] -> Verified & Graded Sources
         - Evidence hierarchy grading (Level I-VII)
         - Predatory journal screening
         - Conflict-of-interest flagging
         - Currency assessment (publication date relevance)
         - Source quality matrix
     |
=== Phase 3: ANALYSIS ===
     |
     |-> [synthesis_agent] -> Synthesis Narrative + Gap Analysis
     |   - Thematic synthesis across sources
     |   - Contradiction identification & resolution
     |   - Evidence convergence/divergence mapping
     |   - Knowledge gap analysis
     |   - Theoretical framework integration
     |
     +-> [devils_advocate_agent] -- CHECKPOINT 2
         - Cherry-picking check
         - Confirmation bias detection
         - Logic chain validation
         - Alternative explanations explored?
         - Verdict: PASS / REVISE
     |
=== Phase 4: COMPOSITION ===
     |
     +-> [report_compiler_agent] -> Full APA 7.0 Draft
         - Title Page
         - Abstract (150-250 words)
         - Introduction (context, problem, purpose, RQ)
         - Literature Review / Theoretical Framework
         - Methodology
         - Findings / Results
         - Discussion (interpretation, implications, limitations)
         - Conclusion & Recommendations
         - References (APA 7.0)
         - Appendices (if applicable)
     |
=== Phase 5: REVIEW (Parallel) ===
     |
     |-> [editor_in_chief_agent] -> Editorial Verdict + Line Feedback
     |   - Originality assessment
     |   - Methodological rigor
     |   - Evidence sufficiency
     |   - Argument coherence
     |   - Writing quality (clarity, conciseness, flow)
     |   - Verdict: ACCEPT / MINOR REVISION / MAJOR REVISION / REJECT
     |
     |-> [ethics_review_agent] -> Research-Integrity Review + Human-Subjects Administrative Status
     |   - AI disclosure compliance
     |   - Attribution integrity
     |   - Dual-use screening
     |   - Fair representation check
     |   - Integrity verdict only: CLEARED / CONDITIONAL / BLOCKED
     |   - Human subjects: readiness and authorization reported separately; institutional determination required
     |   - Authority-bound planning: exact requirement IDs + actor/consumer scope only after the  replay-validated resolved-context gate
     |   - Candidate rule trace: display only a replay-validated and surface-linted  artifact; never use it as a pathway result or workflow input
     |   - Packet structure: consume only a replay-validated  manifest; deterministic status never becomes authorization or content adequacy
     |   - Content coverage: consume only a replay-validated  `LLM-ADVISORY`; preserve deterministic status and report efficacy as `UNMEASURED`
     |
     +-> [devils_advocate_agent] -- CHECKPOINT 3
         - Final vulnerability scan
         - Strongest counter-argument test
         - "So what?" significance check
         - Verdict: PASS / REVISE
     |
=== Phase 6: REVISION ===
     |
     +-> [report_compiler_agent] -> Final Report
         - Address editorial feedback
         - Resolve ethics conditions
         - Incorporate devil's advocate insights
         - Max 2 revision loops
         - Remaining issues -> "Acknowledged Limitations" section

Checkpoint Rules

1. ⚠️ IRON RULE: Devil's Advocate has 3 mandatory checkpoints; Critical-severity issues block progression
2. Revision loops capped at 2 iterations; remaining issues become "acknowledged limitations"
3. ⚠️ IRON RULE: Ethics Review stops the user once to confirm a Critical integrity concern (fabrication / plagiarism / missing AI disclosure / source misrepresentation / concrete harm-enabling specifics). Overridable with recorded reasoning — it confirms, it does not veto. Subject matter alone never blocks; dual-use is advisory (Responsible Use Statement), not a block.
4. User confirmation required after Phase 1 before proceeding

Invocation Modes

This skill runs in 6 phases. Two invocation modes:

Mode A — orchestrator-driven (default): the pipeline_orchestrator_agent (in the ars-pipeline-orchestrator skill) runs all phases end-to-end with state tracking via Material Passport.

Mode B — phase-by-phase (cross-session resume): the user invokes one agent per phase across sessions for long-running projects. To resume in a fresh session, hand the new session the saved phase artifacts (RQ Brief, Methodology Blueprint, source corpus, passport if present) and name the phase to continue from (see ars-pipeline-orchestrator/references/guides/passport_as_reset_boundary.md).

In Mode B, single-phase agents stay strictly within their assigned phase for writes. Reads from upstream phases are allowed. Multi-phase agents (devils_advocate_agent, report_compiler_agent) do exactly the work specified by the caller's invocation for that phase — no extension to other phases in the same call.

Routing into Mode B requires an explicit user signal — naming the mode (e.g. "run fact-check mode") or a [direct-mode] prefix. Ambiguous cross-phase input defaults to clarification per references/shared/intent_clarification_protocol.md.

Enforcement: Phase Boundary blocks on single-phase agents; the multi-phase envelope remains forward-scope. Where the runtime supports automated pre-tool checks they may reinforce this boundary; otherwise the orchestrator enforces it by instruction.

Socratic Mode: Guided Research Dialogue

5-layer dialogue guiding users from vague ideas to concrete research questions. Core principle while non-generation Socratic mode is active: ⚠️ IRON RULE: Never give direct answers. The explicit candidate-generation exit below leaves that mode before any candidate is shown.

Layers: Clarification -> Assumption Probing -> Evidence/Reasoning -> Viewpoint/Perspective -> Implication/Consequence

Research-question authorship boundary: Socratic mode is non-generation by default. Non-convergence may produce only a summary of directions the user has already expressed plus focused questions or a lit-review suggestion; it never produces candidate RQs automatically. If the user explicitly asks the system to propose candidates, announce the exit from non-generation Socratic mode and emit [SOCRATIC-NON-GENERATION-EXIT: explicit_user_request] on a standalone line before any clearly labeled AI-generated candidate. Never switch silently.

See references/guides/socratic_mode_protocol.md for the full 5-layer dialogue flow, management rules, and auto-end conditions.

Opt-in Reading Probe (v3.5.1)

Enabling the ARS_SOCRATIC_READING_PROBE option enables a one-time honesty probe during goal-oriented Socratic sessions. When the user cites a specific paper, the Mentor asks them to paraphrase one passage. Decline is logged without penalty. Default OFF. See references/socratic_mentor_agent.md §"Optional Reading Probe Layer".

Systematic Review Mode

PRISMA 2020-compliant systematic review with optional meta-analysis. Follows 5-phase protocol: Protocol Registration -> Systematic Search -> Screening & Selection -> Data Extraction & RoB -> Synthesis & Reporting.

v3.4.0 compliance: systematic-review mode triggers compliance_agent at Stage 2.5 (Methods items) and Stage 4.5 (remaining items + RAISE 8-role matrix). PRISMA-trAIce Mandatory failures block the pipeline. See references/shared/compliance_checkpoint_protocol.md.

See references/guides/systematic_review_protocol.md for full PRISMA pipeline, checkpoint rules, and meta-analysis procedures.

Operational Modes

Three-Way Scan Mode (WHY / HOW / WHAT)

Use three-way-scan when the user needs a disciplined shortlist of papers compared in a stable frame, but does not yet need a full literature review report.

• WHY: what problem or bottleneck the paper addresses and why it matters
• HOW: what strategy, method, or technical route the paper uses
• WHAT: what the paper found, built, or still leaves unresolved

This mode is intentionally lighter than lit-review. It prioritizes:

1. candidate retrieval
2. deduplication
3. compact per-paper extraction
4. cross-paper synthesis of shared WHY, divergent HOW, and remaining gaps

Recommended per-paper output:

## <paper title>
Source: <provider> | Year: <year> | Link: <url>

- WHY: ...
- HOW: ...
- WHAT: ...

Then add:

• common WHY
• divergent HOW
• strongest WHAT
• unresolved global gap

If the user later wants a broader evidence matrix, thematic synthesis, or PRISMA-like coverage, escalate from three-way-scan to lit-review or systematic-review.

Failure Paths

See references/guides/failure_paths.md for all failure scenarios, trigger conditions, and recovery strategies across all modes.

Key failure path summary:

Literature Monitoring (Optional Post-Pipeline)

Optional post-research monitoring for new publications in the research area.

See references/guides/literature_monitoring_strategies.md for setup instructions across academic databases.

Handoff Protocol: ars-deep-research → ars-academic-paper

After research is complete, the following materials can be handed off to ars-academic-paper:

1. Research Question Brief (from research_question_agent)
2. Methodology Blueprint (from research_architect_agent)
3. Annotated Bibliography (from bibliography_agent)
4. Synthesis Report (from synthesis_agent)
5. [If socratic mode] INSIGHT Collection and Research Plan Summary
6. Preregistration handoff — exactly one builder-produced preregistration-artifact/1.0 sidecar (including an unavailable receipt) and, when status=provided, its explicitly named companion bytes

Trigger: User says "now help me write a paper" or "write a paper based on this"

ars-academic-paper's intake_agent will automatically detect available materials and skip redundant steps:

• Has RQ Brief -> skip topic scoping
• Has Bibliography -> skip literature search
• Has Synthesis -> accelerate findings / discussion writing
• Has preregistration sidecar -> strict-validate it and its named companion, then carry both byte-for-byte; never rebuild it from prose or a template

The non-shell research_architect_agent supplies only the explicit caller declaration and companion handle. Before handoff, a shell-capable dispatcher must run the named deterministic build-preregistration-artifact subcommand in 自动校验脚本（移植版未附带，以文字规则为准）, with caller-held RFC3339 declared_at. Only that builder may create or update the sidecar. A later explicit user supply creates a new builder-produced sidecar; omission or silent substitution is invalid.

See references/examples/handoff_to_paper.md for a detailed handoff example.

Full Academic Pipeline

See ars-pipeline-orchestrator/SKILL.md for the complete workflow.

Agent File References

Reference Files

Templates

Examples

Output Language

Follows the user's language. Academic terminology kept in English. Socratic mode uses natural conversational style.

Anti-Patterns

Explicit prohibitions to prevent common failure modes:

Quality Standards

1. ⚠️ IRON RULE: Every claim must have a citation — no unsupported assertions
2. Evidence hierarchy — meta-analyses > RCTs > cohort studies > case reports > expert opinion (field-neutral baseline; grading is discipline-relative — a source meeting its own field's gold standard can reach Grade A even at a low design level. See references/guides/source_quality_hierarchy.md §Grading Rubric + §Field-Specific Adjustments)
3. Contradiction disclosure — if sources disagree, report both sides with evidence quality comparison
4. Limitation transparency — every report must have an explicit limitations section
5. AI disclosure — all reports include a statement that AI-assisted research tools were used
6. Reproducibility — search strategies, inclusion criteria, and analytical methods must be documented for replication
7. Socratic integrity — while non-generation Socratic mode is active, never give direct answers; always guide through questions. A candidate response is lawful only after the explicit exit marker and is outside that mode.

Cross-Agent Quality Alignment

Unified definitions across all agents. ⚠️ IRON RULE: CRITICAL severity = issue that would invalidate a core conclusion or constitute academic misconduct. Requires immediate resolution.

See references/guides/cross_agent_quality_definitions.md for full peer-reviewed source tiers, currency standards, and severity definitions.

Integration with Other Skills

This skill is domain-agnostic but can be combined with domain-specific skills:

ars-deep-research + tw-hei-intelligence     -> Evidence-based HEI policy research
ars-deep-research + report-to-website       -> Interactive research report
ars-deep-research + podcast-script-generator -> Research podcast
ars-deep-research + ars-academic-paper          -> Full research-to-publication pipeline
ars-deep-research (socratic) + ars-academic-paper (plan) -> Guided research + paper planning
ars-deep-research (systematic-review) + ars-academic-paper -> PRISMA systematic review paper

Model Tiering (optional)

If the dispatching runtime supports per-agent model selection, the session may route this skill's agents per references/shared/model_tiering.md (canonical: the full judgment/execution agent table + rules). Compact rule:

• Unset (default): every agent inherits the session model — behavior identical to tiering disabled.
• economy (frontier-tier session): execution-type agents dispatch ONE tier below the session model — floor high-tier, never lower; judgment-type agents stay on the session model. No-op at or below the floor (announce once).
• quality-boost (below-frontier session): judgment-type agents at the checkpoint surfaces (Stage 2.5/4.5 gates; the opt-in Stage 4→5 claim–ref audit; final review) jump UP to the frontier tier (however many tiers away — not a single increment); nothing is ever downgraded. No-op at the frontier (announce once).
• Unknown values → warn once, behave as unset. Tiers are relative positions, never hard-pinned model ids. When a direction is active, route repeated same-stage calls to the SAME worker so its prompt cache accumulates; unset means dispatch shapes stay byte-equivalent too.

Version Info