agentleFS
justinjohnson25600justinjohnson25600/hermes_skills/client-literature-monitor/SKILL.md

client-literature-monitor

Skill14 starsChanged 26 days agoMITView on GitHub

Use when a business client's case files (flagged analytes, topics, keywords) should drive a recurring literature sweep — PubMed per client, new-items-only digests, silent when nothing new. Built for one metabolic-health practice (name withheld); generalisable to any client with structured case data.

The file

We split this skill into the rules it gives an agent

2 rules. About 0 are common, 2 are distinctive, and 0 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.

  • 0

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    Create the cron job (weekly, agent curates.

    line 37
  2. Distinctive:

    New-only: every PMID is reported once.

    line 51

≈ “Common” is an estimate: at least 1 in 5 of the 372,429 skill 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
SKILL.md

client-literature-monitor — case files drive the literature sweep

Reads each client's structured case data, derives search queries from the items that actually matter (flagged analytes, open topics), sweeps PubMed, and reports only NEW items. Empty result = {"wakeAgent": false} so the cron job stays silent. State is a plain JSON seen-set beside the case files.

When to Use

  • A client engagement produces case records with flagged/abnormal markers (Theorem Metabolic: hormone panels with out-of-range analytes).
  • The client (or you) should know when new published evidence touches their flags.
  • Extending an existing client box: point CLIENTS_ROOT at that box's case directory.

Setup on a client box

  1. Install the skill: python3 install.py client-literature-monitor
  2. Copy scripts/theorem_literature_monitor.py into <HERMES_HOME>/scripts/ of the client profile (cron script: jobs may only reference files under that dir).
  3. Edit the constants at the top of the script for the deployment:
    • CLIENTS_ROOT — the client case directory on this box
    • STATE — where the seen-PMID set lives (beside case data, one per box)
    • QUERY_HINTS — per-analyte query fragments (fallback: quoted phrase search)
  4. Create the cron job (weekly, agent curates; script pre-check stays silent when nothing new):
hermes cron add "30 7 * * 1" "<curation prompt — see scripts/cron_prompt_template.md>" \
  --name "Literature monitor <client>" \
  --script <profile-relative script filename> \
  --deliver local

Contracts (from CAPA-031's detector rules)

  • Sources are honest: a network/API failure prints SOURCES UNKNOWN — never reported as "no new literature".
  • New-only: every PMID is reported once; the seen-set is the dedupe authority.
  • The script never emails anyone. The agent turn decides relevance and delivery; the script is a read-only collector.
  • Case DBs are opened read-only (mode=ro URI).

Reference deployment

a metabolic-health client (name withheld) on the controller box; the weekly cron job (Mondays 07:30) digests to the owner's digest address (configured in cron, not in this repo). First sweep found 15 relevant papers for the first client's flagged analyte; subsequent runs silent until new publications land.

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.