Autonomous Researcher
SkillDocs & knowledgeLets your agent run an end-to-end research project: searching papers, forming hypotheses, and writing a cited report.
Use Autonomous Researcher in Claude, ChatGPT or Ahel Desktop
Free. Sign in, add Autonomous Researcher and connect your AI. About a minute.
Also: Claude Code · Cursor · Codex
Then ask your AI: use the Autonomous Researcher skill
Details
Instructions available. Your AI can read the instructions. Execution depends on the setup they require.
Account requirements not reviewed. Check the skill instructions before use; Ahel provides instructions and does not run this skill.
No other account needed.
Add Ahel to your AI once: Claude, ChatGPT, Cursor, Claude Code or Codex. Then ask it to use this.
About this skill
Self-hosted AI workspace where chat becomes visual workflows, multi-agent operations, and reviewable automations. Local memory; local or cloud models
What this skill tells your AI
The instructions your AI receives, as published by aaronnat23/disp8ch in skills/autonomous-researcher/SKILL.md and read by Ahel’s review.
End-to-end research pipeline: scoping, literature review, hypothesis formation, synthesis, empirical validation, and written output.
Research Pipeline
- Phase 1 — Scope: Decompose the research topic into 3–5 specific sub-questions before starting any search.
- Phase 2 — Literature: Search arXiv, Semantic Scholar, and web sources. Screen for relevance. Target 10–20 sources for a thorough review.
- Phase 3 — Synthesis: Cluster findings by theme. Identify consensus positions, open debates, and knowledge gaps.
- Phase 4 — Hypothesis: Propose 2–3 candidate hypotheses that address a gap. Evaluate feasibility and novelty.
- Phase 5 — Validation (optional): Use the Experiment Loop skill to empirically validate the winning hypothesis — implement, benchmark, keep/discard via git.
- Phase 6 — Output: Write a structured report with: Abstract, Background, Findings, Hypothesis, Experimental Results, Next Steps, References.
Rules
- Use the
councilnode or parallel agents to run competing hypothesis evaluations when multiple stances are plausible. - Store all intermediate findings in memory with appropriate types (fact, event, profile) so subsequent runs can build on prior work.
- Quality gates: after literature review, verify ≥5 sources found before continuing. After synthesis, verify hypothesis addresses a real gap.
- Cite every factual claim. Use inline references
[Author, Year]throughout the report body. - Produce a final
paper_draft.mdartifact saved viawrite_filetodata/workspace/reports/. - When the hypothesis involves code changes or performance optimization, combine with the Experiment Loop skill to move from theory to empirical proof.
API Sources
- arXiv:
https://export.arxiv.org/api/query?search_query={{topic}}&max_results=10 - Semantic Scholar:
https://api.semanticscholar.org/graph/v1/paper/search?query={{topic}}&limit=10&fields=title,abstract,authors,year - CrossRef:
https://api.crossref.org/works?query={{topic}}&rows=10 - DuckDuckGo: use
web_searchtool for general/current web sources
Signals
- GitHub stars
- 100
- Forks
- 2
- Last commit
- Sep 2026
Advanced
- Item type
- skill
- Key
autonomous-researcher- Source
- github.com/aaronnat23/disp8ch
Related picks
Skill · davila7
The pick for Searchsummarize
Skill · zhouguoqing
The pick for Summarizelark-markdown
Skill · larksuite
The pick for Markdownmarkdown-formatter
Skill · nvidia
The pick for Markdownacademic-paper-composer
Skill · brycewang-stanford
The pick for Academic03-academic-writing
Skill · 24kchengye
The pick for Academic