flonat-research
AGENTS.md150 starsChanged 9 days agoMITView on GitHub
15 neutral agent definitions with client-specific adapters. Compatible agents render to Claude Markdown and Codex TOML. Unlike skills, agents solve the "grading your own homework" problem — they review work without access to the author's reasoning. Review agents never modify the author's files. They produce reports and recommendations. The fixer agent is the only exception — it reads a critic report and applies fixes, but only when explicitly invoked. Create a .md file in agents/ with YAML frontmatter: Declare clients…
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 31,774 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 repository89 other files it gives its agents.
- AGENTS.mdAGENTS.md
- CLAUDE.mdCLAUDE.md
- beamer-deckSkill
- bib-coverageSkill
- bib-filterSkill
- bib-parseSkill
- brief-compliance-checkSkill
- camera-readySkill
- causal-designSkill
- checkpointSkill
- code-archaeologySkill
- compile-knowledgeSkill
- computational-experimentsSkill
- cross-language-checkSkill
- data-analysisSkill
- devils-advocateSkill
- docs-consistencySkill
- docxSkill
- experiment-designSkill
- grill-meSkill
- handoffSkill
- ideasSkill
- init-project-courseSkill
- init-project-lightSkill
- init-project-orchestrationSkill
- init-projectSkill
- insights-deckSkill
- interview-meSkill
- knowledge-lintSkill
- latex-diffSkill
- latex-health-checkSkill
- latex-polishSkill
- latex-postersSkill
- latex-scaffoldSkill
- latexSkill
- latex-templateSkill
- lean-checkSkill
- math-proofSkill
- mcp-builderSkill
- method-auditSkill
- multi-perspectiveSkill
- numerical-checkSkill
- pdfSkill
- pipeline-manifestSkill
- playwright-cliSkill
- postmortemSkill
- pre-commit-auditSkill
- pre-submission-reportSkill
- project-deckSkill
- project-safetySkill
- proof-readabilitySkill
- proofreadSkill
- python-envSkill
- quarto-deckSkill
- reorgSkill
- replication-auditSkill
- replication-packageSkill
- retarget-journalSkill
- review-clusterSkill
- review-responseSkill
- The first 60 shown.
The file
We split this agents.md into the rules it gives an agent
1 rule. About 0 are common, 0 are distinctive, and 1 is very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈0
Distinctive · estimate
This author's own choices. Worth a look.
1
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Very specific: line 44
Create a .md file in agents/ with YAML frontmatter:
≈ “Common” is an estimate: at least 1 in 5 of the 40,246 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
Agents
15 neutral agent definitions with client-specific adapters.
Compatible agents render to Claude Markdown and Codex TOML. Unlike skills, agents solve the "grading your own homework" problem — they review work without access to the author's reasoning.
Available Agents
| Agent | Description |
artifact-coherence-auditor | Audits coherence between paper prose and replication outputs — catches\ \ hallucinated results, missing scripts, mismatched numbers, and unverifiable claims.\ \ Read-only with respect to project files; writes its own report at reviews/<scope>/artifact-coherence-auditor/<YYYY-MM-DD-HHMM>.md.\ \ Complements code-paper-auditor (which maps numbers to code) by checking whether\ \ the replication package actually produces what the paper claims |
blindspot | Peripheral vision audit for empirical output |
claim-verify | Verify that cited claims in a paper accurately represent what the source\ \ papers actually say |
code-paper-auditor | Use this agent when you need to verify code-paper consistency — mapping\ \ every quantitative claim in a paper to its source code and output files |
code-review | Multi-persona orchestrator for adversarial review of R, Python, Julia,\ \ or Stata research scripts |
codex-research | Code review and research agent that delegates to OpenAI Codex CLI in\ \ headless mode |
domain-reviewer | Research-focused substantive correctness agent |
fatal-error-check | Fast pre-review check for fatal errors in LaTeX papers |
fixer | Generic fix implementer for any critic report |
gemini-research | Web research agent that delegates to Gemini 4 Argon through the Antigravity\ \ CLI (agy) in headless mode |
paper-critic | Adversarial auditor for LaTeX papers |
peer-reviewer | Use this agent when you need to review someone else's paper — as a\ \ peer reviewer, discussant, or for reading group preparation |
proposal-reviewer | Use this agent when you need to review a research proposal, extended\ \ abstract, conference submission outline, or pre-paper plan — either his own or\ \ someone else's |
referee2-reviewer | Rigorous adversarial reviewer for papers, manuscripts, research designs, code, and arguments |
reproducibility-auditor | Reviews research workflows for reproducibility gaps — hidden dependencies,\ \ absolute paths, undocumented prerequisites, environment assumptions, and output\ \ traceability |
How Agents Work
- The main session selects an agent supported by its client
- The agent runs in a separate context with its own instructions
- The agent produces a report (never modifies source files)
- The report is returned to the main session for review
The Read-Only Principle
Review agents never modify the author's files. They produce reports and recommendations.
The fixer agent is the only exception — it reads a critic report and applies fixes,
but only when explicitly invoked.
Creating a New Agent
Create a .md file in agents/ with YAML frontmatter:
---
name: my-agent
description: "What this agent does"
---Declare clients and required capabilities before rendering. codex-research remains Claude-only because it delegates externally to Codex CLI.
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.