airesearch

AIRESEARCH — AI-Assisted Research

Capability overview

Code: AIRESEARCH Full name: AI-Assisted Research Description: AI-assisted secondary research with source triangulation, citation chains, and confidence flags.

---

Skills index

#	Skill	Command	Purpose
1	Topic Scoping	/cps-airesearch:topic-scope	Frame a research question into PICO / SPIDER style sub-questions and search terms.
2	Source Pull	/cps-airesearch:source-pull	Run multi-source search and pull primary references with citation metadata.
3	Triangulate Findings	/cps-airesearch:triangulate	Cross-check claims across ≥3 independent sources; flag single-source claims.
4	Research Synthesis	/cps-airesearch:synthesis	Synthesize findings into a structured brief with explicit confidence per claim.
5	Research Gap List	/cps-airesearch:gap-list	Identify unresolved questions and recommended primary research.

---

Skill 1: Topic Scoping

Command: /cps-airesearch:topic-scope

Purpose

Frame a research question into PICO / SPIDER style sub-questions and search terms.

Inputs required

client:
  name: "Client Name"
  industry: "Industry"
context:
  scope: "in-scope description"
  constraints: ["constraint 1", "constraint 2"]
references:
  - "Prior deliverable / document name"
  - "External benchmark / source"

Methodology

1. Frame. Translate client request into a structured problem statement and hypotheses.
2. Gather. Pull data from engagement brain, client documents, and external sources.
3. Analyze. Apply the Topic Scoping method using domain frameworks (see Frameworks section of the agent).
4. Synthesize. Produce primary deliverable with recommendations, evidence, and risks.
5. Validate. Run /cps:verify-quality and pass through Every claim has ≥1 cited source with date and author.

Output shape

A CPS-branded Topic Scoping deliverable in 05_Deliverables_Final/. Pyramid-Principle structured, sourced, and reviewed.

Quality checklist

• Every claim has ≥1 cited source with date and author
• Single-source claims are flagged "single-source — verify"
• Confidence labels (high/medium/low) are explicit on every conclusion

---

Skill 2: Source Pull

Command: /cps-airesearch:source-pull

Purpose

Run multi-source search and pull primary references with citation metadata.

Inputs required

client:
  name: "Client Name"
  industry: "Industry"
context:
  scope: "in-scope description"
  constraints: ["constraint 1", "constraint 2"]
references:
  - "Prior deliverable / document name"
  - "External benchmark / source"

Methodology

1. Frame. Translate client request into a structured problem statement and hypotheses.
2. Gather. Pull data from engagement brain, client documents, and external sources.
3. Analyze. Apply the Source Pull method using domain frameworks (see Frameworks section of the agent).
4. Synthesize. Produce primary deliverable with recommendations, evidence, and risks.
5. Validate. Run /cps:verify-quality and pass through Every claim has ≥1 cited source with date and author.

Output shape

A CPS-branded Source Pull deliverable in 05_Deliverables_Final/. Pyramid-Principle structured, sourced, and reviewed.

Quality checklist

• Every claim has ≥1 cited source with date and author
• Single-source claims are flagged "single-source — verify"
• Confidence labels (high/medium/low) are explicit on every conclusion

---

Skill 3: Triangulate Findings

Command: /cps-airesearch:triangulate

Purpose

Cross-check claims across ≥3 independent sources; flag single-source claims.

Inputs required

client:
  name: "Client Name"
  industry: "Industry"
context:
  scope: "in-scope description"
  constraints: ["constraint 1", "constraint 2"]
references:
  - "Prior deliverable / document name"
  - "External benchmark / source"

Methodology

1. Frame. Translate client request into a structured problem statement and hypotheses.
2. Gather. Pull data from engagement brain, client documents, and external sources.
3. Analyze. Apply the Triangulate Findings method using domain frameworks (see Frameworks section of the agent).
4. Synthesize. Produce primary deliverable with recommendations, evidence, and risks.
5. Validate. Run /cps:verify-quality and pass through Every claim has ≥1 cited source with date and author.

Output shape

A CPS-branded Triangulate Findings deliverable in 05_Deliverables_Final/. Pyramid-Principle structured, sourced, and reviewed.

Quality checklist

• Every claim has ≥1 cited source with date and author
• Single-source claims are flagged "single-source — verify"
• Confidence labels (high/medium/low) are explicit on every conclusion

---

Skill 4: Research Synthesis

Command: /cps-airesearch:synthesis

Purpose

Synthesize findings into a structured brief with explicit confidence per claim.

Inputs required

client:
  name: "Client Name"
  industry: "Industry"
context:
  scope: "in-scope description"
  constraints: ["constraint 1", "constraint 2"]
references:
  - "Prior deliverable / document name"
  - "External benchmark / source"

Methodology

1. Frame. Translate client request into a structured problem statement and hypotheses.
2. Gather. Pull data from engagement brain, client documents, and external sources.
3. Analyze. Apply the Research Synthesis method using domain frameworks (see Frameworks section of the agent).
4. Synthesize. Produce primary deliverable with recommendations, evidence, and risks.
5. Validate. Run /cps:verify-quality and pass through Every claim has ≥1 cited source with date and author.

Output shape

A CPS-branded Research Synthesis deliverable in 05_Deliverables_Final/. Pyramid-Principle structured, sourced, and reviewed.

Quality checklist

• Every claim has ≥1 cited source with date and author
• Single-source claims are flagged "single-source — verify"
• Confidence labels (high/medium/low) are explicit on every conclusion

---

Skill 5: Research Gap List

Command: /cps-airesearch:gap-list

Purpose

Identify unresolved questions and recommended primary research.

Inputs required

client:
  name: "Client Name"
  industry: "Industry"
context:
  scope: "in-scope description"
  constraints: ["constraint 1", "constraint 2"]
references:
  - "Prior deliverable / document name"
  - "External benchmark / source"

Methodology

1. Frame. Translate client request into a structured problem statement and hypotheses.
2. Gather. Pull data from engagement brain, client documents, and external sources.
3. Analyze. Apply the Research Gap List method using domain frameworks (see Frameworks section of the agent).
4. Synthesize. Produce primary deliverable with recommendations, evidence, and risks.
5. Validate. Run /cps:verify-quality and pass through Every claim has ≥1 cited source with date and author.

Output shape

A CPS-branded Research Gap List deliverable in 05_Deliverables_Final/. Pyramid-Principle structured, sourced, and reviewed.

Quality checklist

• Every claim has ≥1 cited source with date and author
• Single-source claims are flagged "single-source — verify"
• Confidence labels (high/medium/low) are explicit on every conclusion

---

Cross-skill workflows

Discovery → /cps-airesearch:topic-scope → /cps-airesearch:source-pull → Recommendations

Integration with other CPS plugins

Plugin	Integration point
cps-ai-regs	Regulatory research with freshness gates
cps-ai-dd	Pre-DD market and target research
cps-str	Strategy research feeding into market sizing

Templates & artifacts

All deliverables use standard CPS branding via:

• /doc-gen for document generation
• assets/cps-branding.json for styling
• scripts/cps-document-generator.js for automation

Skill	Primary artifact	Format
/cps-airesearch:topic-scope	Topic Scoping Report	DOCX/PDF
/cps-airesearch:source-pull	Source Pull Report	DOCX/PDF
/cps-airesearch:triangulate	Triangulate Findings Report	DOCX/PDF
/cps-airesearch:synthesis	Research Synthesis Report	DOCX/PDF
/cps-airesearch:gap-list	Research Gap List Report	DOCX/PDF