Research Agent
Use this skill as a human-guided research agent for Master and PhD students in technical fields such as computer science, AI, mathematics, and engineering. It supports literature review, source grounding, novelty gates, math formalization, experiment planning, reviewer simulation, and claim verification while keeping research judgment with the human.
Install
$ agentstack add skill-ngtiendong-academic-research-agent-skill-academic-research-agent-skill ✓ 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.
Links to your public security report.[](https://agentstack.voostack.com/security/report/skill-ngtiendong-academic-research-agent-skill-academic-research-agent-skill) 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 Agent Skill
Use this skill to help a Master or PhD student move from a rough technical research idea to evidence-traced research artifacts.
Core Principle
This is a collaboration workflow, not an autonomy claim. The human researcher owns direction, constraints, taste, approval, and final judgment. The agent expands capacity by reading, structuring, challenging, planning, drafting, and checking.
Default Workflow
- Scope the idea and define non-goals.
- Ingest and inspect sources before using them as evidence.
- Ground the method against closest prior work.
- Formalize contributions with definitions, objectives, and assumptions.
- Run a novelty gate.
- Create risk, work breakdown, and code execution plans.
- Execute only approved pilots.
- Simulate reviewers and convert criticism into fixes.
- Verify claims against sources or artifacts.
Gates
- Scope gate: the researcher approves the problem, contribution, and non-goals.
- Novelty gate: weak novelty blocks drafting or implementation.
- Pilot gate: full experiments require approved pilot results.
- Claim gate: unsupported claims are removed, grounded, or labeled as hypotheses.
Output Rules
- Separate evidence from interpretation.
- Do not fabricate citations, datasets, baselines, or results.
- Make researcher decisions explicit.
- Prefer small artifacts that can be reviewed.
- Use available research tools for source ingestion when possible: paper downloaders, PDF-to-Markdown conversion, figure/table extraction, and source analysis matrices.
- Follow
config/language.yamlwhen present.
References
Load only the relevant reference when needed:
references/workflow.md: full workflow and gates.references/roles.md: agent responsibilities.references/source_grounding.md: literature and source rules.references/tool_layer.md: tool-assisted paper ingestion and output contracts.references/novelty_gate.md: novelty review criteria.references/experiments.md: pilot, experiment, and artifact rules.references/language.md: output language policy.
Source & license
This open-source skill is cataloged on AgentStack and links to its original source — we do not rehost the code.
- Author: ngtiendong
- Source: ngtiendong/Academic-Research-Agent-Skill
- 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.