Deep Research

Vocabulary

~/.claude/research/

Dispatch

$ARGUMENTS

?

?

check <claim>

verify <claim>

compare <A> vs <B> [vs <C>...]

survey <field or topic>

track <topic>

resume [number or keyword]

list [active|domain|tier]

archive

delete <N>

export [N]

Auto-Detection Heuristic

1. Ends with

   ?

   or starts with question word (who/what/when/where/why/how/is/are/can/does/should/will) → Investigate
2. Contains

   vs

   ,

   versus

   ,

   compared to

   ,

   or

   between noun phrases → Compare
3. Declarative statement with factual claim, no question syntax → Fact-check
4. Broad field name with no specific question → ask: "Investigate a specific question, or survey the entire field?"
5. Ambiguous → ask: "Would you like me to investigate this question, verify this claim, or survey this field?"

Gallery (Empty Arguments)

Pick a number, paste your own question, or type

guide me

.

Skill Awareness

/honest-review

/wargame

/host-panel

/prompt-engineer

Complexity Classification

--depth <tier>

Wave Pipeline

Wave 0: Triage (always inline, never parallelized)

1. Run

   !uv run python skills/research/scripts/research-scanner.py "$ARGUMENTS"

   for deterministic pre-scan
2. Decompose query into 2-5 sub-questions
3. Score complexity on the 5-dimension rubric
4. Check tool availability — probe key MCP tools; set degraded mode flags and confidence ceilings per

   references/source-selection.md

5. Select tools per domain signals — read

   references/source-selection.md

6. Check for existing journals — if

   track

   or

   resume

   , load prior state
7. Present triage to user — show: complexity score, sub-questions, planned strategy, estimated tier. User may override.

Wave 1: Broad Sweep (parallel)

Subagent A → brave-search + duckduckgo-search for sub-question 1
Subagent B → exa + g-search for sub-question 2
Subagent C → context7 / deepwiki / arxiv / semantic-scholar for technical specifics
Subagent D → wikipedia / wikidata for factual grounding
[Subagent E → PubMed / openalex if academic domain detected]

"research-{slug}"

Lead: triage (Wave 0), orchestrate, judge reconcile (Wave 3), synthesize (Wave 4)
  |-- web-researcher:       brave-search, duckduckgo-search, exa, g-search
  |-- tech-researcher:      context7, deepwiki, arxiv, semantic-scholar, package-version
  |-- content-extractor:    fetcher, trafilatura, docling, wikipedia, wayback
  |-- [academic-researcher: arxiv, semantic-scholar, openalex, crossref, PubMed]
  |-- [adversarial-reviewer: devil's advocate — counter-search all emerging findings]

{
  "sub_question": "...",
  "findings": [{"claim": "...", "source_url": "...", "source_tool": "...", "excerpt": "...", "confidence_raw": 0.6}],
  "leads": ["url1", "url2"],
  "gaps": ["could not find data on X"]
}

Wave 1.5: Perspective Expansion (Deep/Exhaustive only)

Wave 2: Deep Dive (parallel, targeted)

1. Rank leads from Wave 1 by potential value (citation frequency, source authority, relevance)
2. Dispatch deep-read subagents — use fetcher/trafilatura/docling to extract full content from top leads
3. Follow citation chains — if a source cites another, fetch the original
4. Fill gaps — for each gap identified in Wave 1, dispatch targeted searches
5. Use thinking MCPs:

   • cascade-thinking

     for multi-perspective analysis of complex findings

   • structured-thinking

     for tracking evidence chains and contradictions

   • think-strategies

     for complex question decomposition (Standard+ only)

Wave 3: Cross-Validation (parallel)

references/confidence-rubric.md

references/self-verification.md

1. Independence check — are supporting sources truly independent? Sources citing each other are NOT independent.
2. Counter-search — explicitly search for evidence AGAINST each major claim using a different search engine
3. Freshness check — verify sources are current (flag if >1 year old for time-sensitive topics)
4. Contradiction scan — read

   references/contradiction-protocol.md

   , identify and classify disagreements
5. Confidence scoring — assign 0.0-1.0 per

   references/confidence-rubric.md

6. Bias sweep — check each finding against 10 bias categories (7 core + 3 LLM-specific) per

   references/bias-detection.md

references/self-verification.md

For each finding, attempt to disprove it. Search for counterarguments. Check if evidence is outdated. Verify claims actually follow from cited evidence. Flag LLM confabulations.

Wave 4: Synthesis (always inline, lead only)

references/output-formats.md

1. Answer directly — answer the user's question clearly
2. Map evidence — all verified findings with confidence and citations
3. Surface contradictions — where sources disagree, with analysis of why
4. Show confidence landscape — what is known confidently, what is uncertain, what is unknown
5. Audit biases — biases detected during research
6. Identify gaps — what evidence is missing, what further research would help
7. Distill takeaways — 3-7 numbered key findings
8. Cite sources — full bibliography with provenance

• Investigate → Research Brief (Standard) or Deep Report (Deep/Exhaustive)
• Fact-check → Quick Answer with verdict + evidence
• Compare → Decision Matrix
• Survey → Annotated Bibliography
• User can override with

  --format brief|deep|bib|matrix

Confidence Scoring

• No claim reported at >= 0.7 unless supported by 2+ independent sources
• Single-source claims cap at 0.6 regardless of source authority
• Degraded mode (all research tools unavailable): max confidence 0.4, all findings labeled "unverified"

c_merged = 1 - (1-c1)(1-c2)...(1-cN)

Evidence Chain Structure

FINDING RR-{seq:03d}: [claim statement]
  CONFIDENCE: [0.0-1.0]
  EVIDENCE:
    1. [source_tool] [url] [access_timestamp] — [relevant excerpt, max 100 words]
    2. [source_tool] [url] [access_timestamp] — [relevant excerpt, max 100 words]
  CROSS-VALIDATION: [agrees|contradicts|partial] across [N] independent sources
  BIAS MARKERS: [none | list of detected biases with category]
  GAPS: [none | what additional evidence would strengthen this finding]

!uv run python skills/research/scripts/finding-formatter.py --format markdown

Source Selection

references/source-selection.md

Bias Detection

references/bias-detection.md

State Management

• Journal path:

  ~/.claude/research/

• Archive path:

  ~/.claude/research/archive/

• Filename convention:

  {YYYY-MM-DD}-{domain}-{slug}.md

  • {domain}

    :

    tech

    ,

    academic

    ,

    market

    ,

    policy

    ,

    factcheck

    ,

    compare

    ,

    survey

    ,

    track

    ,

    general

  • {slug}

    : 3-5 word semantic summary, kebab-case
  • Collision: append

    -v2

    ,

    -v3

• Format: YAML frontmatter + markdown body +

  <!-- STATE -->

  blocks

• Quick: save once at end with

  status: Complete

• Standard/Deep/Exhaustive: save after Wave 1 with

  status: In Progress

  , update after each wave, finalize after synthesis

1. resume

   (no args): find

   status: In Progress

   journals. One → auto-resume. Multiple → show list.

2. resume N

   : Nth journal from

   list

   output (reverse chronological).

3. resume keyword

   : search frontmatter

   query

   and

   domain_tags

   for match.

!uv run python skills/research/scripts/journal-store.py

<!-- STATE
wave_completed: 2
findings_count: 12
leads_pending: ["url1", "url2"]
gaps: ["topic X needs more sources"]
contradictions: 1
next_action: "Wave 3: cross-validate top 8 findings"
-->

In-Session Commands (Deep/Exhaustive)

drill <finding #>

pivot <new angle>

counter <finding #>

export

status

sources

confidence

gaps

?

references/session-commands.md

Reference File Index

references/source-selection.md

references/confidence-rubric.md

references/evidence-chain.md

references/bias-detection.md

references/contradiction-protocol.md

references/self-verification.md

references/output-formats.md

references/team-templates.md

references/session-commands.md

references/dashboard-schema.md

export

Critical Rules

1. No claim >= 0.7 unless supported by 2+ independent sources — single-source claims cap at 0.6
2. Never fabricate citations — if URL, author, title, or date cannot be verified, use vague attribution ("a study in this tradition") rather than inventing specifics
3. Always surface contradictions explicitly — never silently resolve disagreements; present both sides with evidence
4. Always present triage scoring before executing research — user must see and can override complexity tier
5. Save journal after every wave in Deep/Exhaustive mode — enables resume after interruption
6. Never skip Wave 3 (cross-validation) for Standard/Deep/Exhaustive tiers — this is the anti-hallucination mechanism
7. Multi-engine search is mandatory for fact-checking — use minimum 2 different search tools (e.g., brave-search + duckduckgo-search)
8. Apply the Accounting Rule after every parallel dispatch — N dispatched = N accounted for before proceeding to next wave
9. Distinguish facts from interpretations in all output — factual claims carry evidence; interpretive claims are explicitly labeled as analysis
10. Flag all LLM-prior findings — claims matching common training data but lacking fresh evidence must be flagged with bias marker
11. Max confidence 0.4 in degraded mode — when all research tools are unavailable, report all findings as "unverified — based on training knowledge"
12. Load ONE reference file at a time — do not preload all references into context
13. Track mode must load prior journal before searching — avoid re-researching what is already known
14. The synthesis is not a summary — it must integrate findings into novel analysis, identify patterns across sources, and surface emergent insights not present in any single source
15. PreToolUse Edit hook is non-negotiable — the research skill never modifies source files; it only creates/updates journals in

    ~/.claude/research/