autoresearch — Meridian research-loop prompts

Repository-documented prompt package for coordinating AI research loops in the Mars/Meridian ecosystem, with research-tree exploration, investigation design, run supervision and preserved research memory.

Workflow level n/a Source Repository-documented Outcome Not empirically evaluated here

Evidence & verification

Source level

Repository-documented
Grounded in a maintainer repository/README. Availability, stars and forks do not establish research benefit or correctness.

Outcome verification

Not empirically evaluated here
A documented resource or intended workflow, not a demonstrated academic outcome. No empirical benefit is established by its inclusion.

Sources checked

2026-10-17 (UTC)

README-only capability/concept description. No performance outcome is asserted or verified. Adoption guidance is editorial and untested.

Source-checked does not mean independently reproduced. See the Method page for how these labels are assigned.

Sources (1)

Research applicability

Provides a bounded coordination pattern for researchers organizing AI-assisted investigations and experiments; not a claim of autonomous scientific success.

Getting started

Editorial, untested guidance: read the README and inspect skills/autoresearch and the lead/runner prompts before syncing into a sandbox Mars/Meridian project; approve goals and review run outputs.

Maintenance

Mutable current README inspected; ongoing maintenance not independently assessed.

License

Not audited; inspect current license and dependency/data rights before reuse.

Workflow & human oversight

Resource-level automation unspecified. Researchers approve objectives and inspect outputs; documentation is not an executed workflow.

Workflow level is not an evidence score; higher is not better evidence.

Approach

The README separates a lead agent requiring a user-approved goal and design delegation from a runner that follows the approved goal and stage brief without redefining success.

Limitations

README-only source review by an AI editorial agent; no implementation execution, independent outcome evaluation, security audit or full linked-document review. HEAD is mutable; saved content and hashes preserve the inspected version. Same-name autoresearch projects are not presumed identical. GitHub identifies this as haowjy/autoresearch, not a fork; Mars/Meridian are dependencies/ecosystem, not identity aliases.

Reproducibility

Fresh README and identity metadata checked; no software or outcomes reproduced.

Adoption requirements

Mars/Meridian downstream project and the named meridian-dev-workflow/design-lead dependencies; exact installation and compatibility have not been audited.

Comments

No comments yet.

Add a comment

Comments are held for moderation before appearing. Critical or corrective comments are welcome — especially evidence that an entry overstates what has been verified.