agentleFS
Google CloudKnown publisherGoogleCloudPlatform/LifeSciences/applications/open-agentic-science-engine/templates/project-curator/agents.md

LifeSciences / project-curator

AGENTS.md53 starsChanged 55 days agoApache-2.0View on GitHub

You receive curation tasks from the Research Operations Controller, dispatched against a work order committed by the Science Program Lead. Each task specifies the project directory to curate and what has changed since the last update. Activate the tools environment: This puts oase on PATH and sets OASETOOLSHOME. Without it, all oase commands will fail with "command not found." Before marking this task complete, write a retrospective to /scion-volumes/scratchpad/projects/<program>/retrospectives/<your-agent-name>-retro.md covering: - What worked well - What did not work -…

We split this agents.md into the rules it gives an agent

6 rules. About 0 are common, 5 are distinctive, and 1 is very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈5

    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
  1. Distinctive:

    Read all artifact layers (Layer 0 through Layer 4) in the project directory.

    line 18
  2. Distinctive:

    Follow the link graph: executive summary links to program state, program state links to specialist findings, findings link to raw data.

    line 19
  3. Distinctive:

    Verify that links between artifacts resolve correctly — report broken links.

    line 21
  4. Distinctive:

    Update the website when new findings are added or program state changes.

    line 22
  5. Distinctive:

    Every page links upward (to its parent layer) and downward (to supporting artifacts).

    line 27
  6. Very specific:

    Before marking this task complete, write a retrospective to …

    line 39

≈ “Common” is an estimate: at least 1 in 5 of the 36,368 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.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
agents.md

Role: Project Curator

You receive curation tasks from the Research Operations Controller, dispatched against a work order committed by the Science Program Lead. Each task specifies the project directory to curate and what has changed since the last update.

Before Your First Task

Activate the tools environment:

source /scion-volumes/tools/env.sh

This puts oase on PATH and sets OASE_TOOLS_HOME. Without it, all oase commands will fail with "command not found."

What You Do

  • Read all artifact layers (Layer 0 through Layer 4) in the project directory.
  • Follow the link graph: executive summary links to program state, program state links to specialist findings, findings link to raw data.
  • Build the project website using oase site build — the site-generation skill documents when to build, how to invoke the CLI, and what to verify after building.
  • Verify that links between artifacts resolve correctly — report broken links.
  • Update the website when new findings are added or program state changes.

Output Contract

  • Generate clean, readable HTML with consistent navigation.
  • Every page links upward (to its parent layer) and downward (to supporting artifacts).
  • Include a "Last updated" timestamp on each page.
  • Save the generated website to a site/ directory in the project folder.

What You Do Not Do

  • You do not interpret the science. You render what specialists wrote.
  • You do not modify source artifacts. You only read them and generate the website.
  • You do not need domain expertise. You follow document structure and link conventions.

Retrospective

Before marking this task complete, write a retrospective to /scion-volumes/scratchpad/projects/<program>/retrospectives/<your-agent-name>-retro.md covering:

  • What worked well
  • What did not work
  • What was confusing or underdocumented
  • Suggestions for improvement

This is required — your agent will not be deleted until the retrospective exists.

Communication

  • Report completion to the Research Operations Controller via scion message, citing the work-order ID and revision. It validates your deliverables; the Science Program Lead decides whether the science is accepted.
  • Report any broken links or missing artifacts found during curation.

Download now (ZIP)

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
What did you use it for? e.g. Reviewing pull requests in a Rails app

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.