autoresearch-paper
AGENTS.md2 starsChanged 14 days agoMITView on GitHub
Use the main autoresearch skill for research projects. Read .claude/skills/autoresearch/SKILL.md and .claude/commands/autoresearch.md. The same workflow covers intake, venue calibration, research planning, experiments, writing, independent review, and verified delivery. Use autoresearch iterate for a bounded metric loop. Define the goal, edit scope, metric, verification command, and any regression guard. Make one change per iteration. Commit it, verify the metric and guard, keep or revert the change, and log the result. Use autoresearch --auto for predicate orchestration. The command guides list auxiliary…
Where files are headingNew AGENTS.md files have more rules about keeping secrets out of the code, and fewer about code style.
Of the AGENTS.md files we read, the 33,315 first seen in the last six months, against the older ones: the kind of rule that gained most share, and the kind that lost most. About files like this in general, not this one, and an estimate, since a kind is a pattern in the words.
In this repository2 other files it gives its agents.
- autoresearchSkill
- autoresearchSkill
The file
We split this AGENTS.md into the rules it gives an agent
8 rules. About 0 are common, 2 are distinctive, and 6 are very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈2
Distinctive · estimate
This author's own choices. Worth a look.
6
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Very specific: line 3
Use the main autoresearch skill for research projects.
- Very specific: line 10
Use context to load the current stage and its evidence requirements.
- Distinctive: line 11
Ask the user for venue and budget decisions through the host's input tool.
- Distinctive: line 12
Keep frozen research plans, evaluators, and run evidence outside metric-loop rollback.
- Very specific: line 13
Follow foundation/references/runtime-contract.md for run state and recovery.
- Very specific: line 17
Use autoresearch iterate for a bounded metric loop.
- Very specific: line 27
Run bash scripts/transform.sh to update the installed distributions.
- Very specific: line 28
Run bash scripts/test-paper.sh for the research workflow and use the relevant shell suites in tests/ for packaging and engineering commands.
≈ “Common” is an estimate: at least 1 in 5 of the 42,283 AGENTS.md files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
Autoresearch Paper
Use the main autoresearch skill for research projects. Read
.claude/skills/autoresearch/SKILL.md and .claude/commands/autoresearch.md.
The same workflow covers intake, venue calibration, research planning,
experiments, writing, independent review, and verified delivery.
Research runs
- Use
contextto load the current stage and its evidence requirements. - Ask the user for venue and budget decisions through the host's input tool.
- Keep frozen research plans, evaluators, and run evidence outside metric-loop rollback.
- Follow
foundation/references/runtime-contract.mdfor run state and recovery.
Engineering commands
Use autoresearch iterate for a bounded metric loop. Define the goal, edit
scope, metric, verification command, and any regression guard. Make one
change per iteration. Commit it, verify the metric and guard, keep or revert
the change, and log the result. Use autoresearch --auto for predicate
orchestration. The command guides list auxiliary commands.
Maintain the skill
- Edit command and skill instructions in
.claude/. - Edit the research controller, its references, and tests in
foundation/. - Run
bash scripts/transform.shto update the installed distributions. - Run
bash scripts/test-paper.shfor the research workflow and use the relevant shell suites intests/for packaging and engineering commands. - Preserve source attribution and the copyright notices in
LICENSE.
See the installation guide and contribution guide.
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
No reviews yet
People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.
- It worked
- Partly
- It didn't
Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

