Skip to main content

Pool intake

The pool is the evidence base Hermes uses to tailor resumes honestly.

For the full operator walkthrough, use Resume Agent > Pool Content Guide. This page stays brief on purpose and acts as the shared reference entry point.

What goes into the pool​

  • work experience
  • personal projects
  • OSS contributions

Why it matters​

The pipeline cannot produce strong resumes from weak source material. project-selection and point-repointing both depend on the pool being structured, complete, and credible.

How to onboard material​

Use:

  • pool-intake to add new items
  • pool-versioning to understand the structure and write boundaries

Run this after profile-bootstrap has already been used to replace candidate placeholders locally. At that point, pool-intake becomes the normal way to take the user's experience, project, or OSS markdown files and place them into the right location and format on the VPS or wherever Hermes is running.

The standard intake result is:

  • raw.md copied or transformed into the pool item folder
  • idx.md initialized with current_version: 0
  • empty versions/
  • masters.md updated for personal_project and oss_contribution

Quality bar​

The pool should contain enough detail that the system can answer:

  • what was built
  • why it mattered
  • what technologies were actually used
  • what results or outputs are provable

If that evidence is absent, the pipeline should stay conservative rather than inventing claims.

Next reading:

If you want the actual pool-related skill files and file contracts referenced on this page, use the public source repository.

hermes-autonomous-resume on GitHub