Install
$ agentstack add skill-mmthebest-research-architect-research-architect-draft ✓ scanned · ✓ verified, works with Claude Code, Cursor, and more.
Security review
✓ PassedNo issues found. Passed automated security review. · v0.1.0 How review works →
- ✓ Prompt-injection patterns
- ✓ Secret / credential exfiltration
- ✓ Dangerous shell & filesystem operations
- ✓ Untrusted network calls
- ✓ Known-malicious package signatures
What it can access
- ✓ Network access No
- ✓ Filesystem access No
- ✓ Shell / process execution No
- ✓ Environment & secrets No
- ✓ Dynamic code execution No
From automated source analysis of v0.1.0. “Used” means the capability is present in the source — more access means more to trust, not that it’s unsafe.
Verified badge
Passed review? Show it. Paste this badge into your README, it links to the public security report.
Reliability & compatibility
Declared compatibility
Compatibility is declared by the source manifest. End-to-end runtime verification is coming, see below.
We're building live execution health for every listing: tool-call success rate, median latency, uptime, and last-checked timestamps, measured, not self-reported. It isn't live yet, so we don't show numbers we can't stand behind.
How agent discovery & health will work →About
Research Architect Draft
Use this branch after the upstream research-construction artifacts exist.
Blocking inputs
Draft only when these files or compatible equivalents are available:
paper_output/confirmed_research_spine.md
paper_output/evidence_bank.md
paper_output/claim_register.md
paper_output/citation_support_bank.md
paper_output/evidence_display_map.md
paper_output/section_blueprints.md
paper_output/writing_rationale_matrix.md
paper_output/exemplar_adaptation_plan.md # required when target references exist
Read figure_asset_map.md as the compatible specialized equivalent of evidence_display_map.md.
When evidence is incomplete, produce or update the research execution plan.
Required outputs
Create:
paper_output/section_blueprints.md
paper_output/writing_rationale_matrix.md
paper_output/first_draft/main.md
paper_output/final_artifact_manifest.md
Optional when useful:
paper_output/first_draft/main.tex
Draft architecture
Choose a structure that fits the selected design family, research spine, target-paper functions, and evidence mode. Supported architectures include:
- hypothesis-led quantitative article;
- causal or explanatory evaluation;
- qualitative thematic or interpretive article;
- grounded or process-model article;
- single-case or comparative-case article;
- historical or archival argument;
- mixed-methods article;
- conceptual, critical, normative, or theory-building article;
- policy analysis;
- systematic, scoping, or integrative review;
- computational, experimental, or design/build/evaluation article.
The target paper supplies functional lessons. Its exact headings, section proportions, display sequence, and rhetoric transfer only when they fit the user's independent inquiry.
Drafting rules
- The title and abstract compress the confirmed research spine and inference boundary.
- The introduction moves through significance, prior knowledge, unresolved tension, research question, and contribution in the form appropriate to the design family.
- Theory appears in the role declared by the spine and exemplar analysis.
- Methods, sources, or argumentative procedures include enough detail to evaluate the warrant.
- Findings follow the claim-support sequence rather than the chronology of the research process.
- Quantitative results report effect size, uncertainty, and design limits when applicable.
- Qualitative findings preserve context, evidence excerpts, alternative interpretations, and the relationship between data and themes.
- Case and historical findings make chronology, source provenance, and rival explanations visible.
- Mixed-method findings identify integration points and meta-inferences.
- Conceptual and normative arguments expose premises, steps, counterarguments, and applicability limits.
- Discussion distinguishes evidence, interpretation, transfer, and speculation.
- Every central claim maps to evidence IDs, citation IDs, or explicit limitation language.
- Citation support and exemplar learning remain separate.
- Unverified sources remain visibly marked.
Writing rationale matrix
Use:
| Row ID | Manuscript Unit | Planned Function | Research-Spine Link | Reference-Logic Principle Applied | Project-Specific Adaptation | Evidence or Citation Anchor | Planned Text Move | Claim Boundary | Final Check | |---|---|---|---|---|---|---|---|---|---|
Write the matrix before or during drafting. Each reference-derived move should identify its research function and project-specific adaptation.
Output behavior
The first draft may remain rough while retaining logical completeness. Mark uncertain claims with TODO_VERIFY, missing evidence with TODO_EVIDENCE, adaptation decisions with TODO_ADAPT, and citation gaps with TODO_CITATION.
Source & license
This open-source skill is cataloged on AgentStack and links to its original source — we do not rehost the code.
- Author: mmTheBest
- Source: mmTheBest/Research-Architect
- License: MIT
Install and usage instructions live in the source repository linked above.
Reviews
No reviews yet, be the first.
Write a review
Versions
- v0.1.0 Imported from the upstream source.