Files
hermes/optional-skills/devops/watchers/SKILL.md
Teknium ea8e608821 feat(skills): watchers skill — poll RSS / HTTP JSON / GitHub via cron no-agent (#21881)
* feat(skills): watchers skill — poll RSS / HTTP JSON / GitHub via cron no-agent

Ships three reusable polling scripts plus a shared watermark helper as an
optional skill.  Users wire them into the existing cron (no_agent=True)
mode rather than learning a new subsystem.

Supersedes the closed PR #21497 (parallel watcher subsystem).  Same value,
zero new core surface.

## What ships

- optional-skills/devops/watchers/SKILL.md: pattern + three example cron commands
- optional-skills/devops/watchers/scripts/_watermark.py: shared helper
  (atomic state writes, bounded ID set, first-run baseline)
- optional-skills/devops/watchers/scripts/watch_rss.py: RSS 2.0 + Atom
- optional-skills/devops/watchers/scripts/watch_http_json.py: any JSON endpoint
  with configurable id_field / items_path / headers
- optional-skills/devops/watchers/scripts/watch_github.py: issues / pulls /
  releases / commits (uses GITHUB_TOKEN if present)

## Invariants enforced by the shared helper

- First run records baseline, emits nothing (never replays existing feed)
- Watermark file is <state_dir>/<name>.json, atomic replace on write
- Bounded to 500 IDs (configurable)
- Empty stdout when no new items — cron treats that as silent delivery

## Validation
- watch_rss.py against news.ycombinator.com/rss first run → empty stdout, watermark populated
- Removed one seen-id, second run → emitted exactly that item
- No DeprecationWarnings (ET element truth-value footgun dodged explicitly)

End-user pattern: 'hermes cron create my-feed --schedule "*/15 * * * *" --no-agent --script $HERMES_HOME/skills/devops/watchers/scripts/watch_rss.py --script-args "--name hn --url https://news.ycombinator.com/rss" --deliver telegram'

* docs(skills/watchers): tighten description to match peer optional skills

* docs(skills/watchers): align frontmatter + structure with peer optional skills

* docs(skills/watchers): gate to linux/macos (shell syntax in examples)
2026-05-08 09:27:15 -07:00

4.4 KiB

name, description, version, author, license, platforms, metadata
name description version author license platforms metadata
watchers Poll RSS, JSON APIs, and GitHub with watermark dedup. 1.0.0 Hermes Agent MIT
linux
macos
hermes
tags category requires_toolsets related_skills
cron
polling
rss
github
http
automation
monitoring
devops
terminal

Watchers

Poll external sources on an interval and react only to new items. Three ready-made scripts plus a shared watermark helper; wire them into a cron job (or run them ad-hoc from the terminal).

When to Use

  • User wants to watch an RSS/Atom feed and be notified of new entries
  • User wants to watch a GitHub repo's issues / pulls / releases / commits
  • User wants to poll an arbitrary JSON endpoint and get notified on new items
  • User asks for "a watcher for X" or "notify me when X changes"

Mental model

A watcher is just a script that:

  1. Fetches data from the external source
  2. Compares against a watermark file of previously-seen IDs
  3. Writes the new watermark back
  4. Prints new items to stdout (or nothing on no-change)

The scripts below handle all three. The agent runs them via the terminal tool — from a cron job, a webhook, or an interactive chat — and reports what's new.

Ready-made scripts

All three live in $HERMES_HOME/skills/devops/watchers/scripts/ once the skill is installed. Each reads WATCHER_STATE_DIR (defaults to $HERMES_HOME/watcher-state/) for its state file, keyed by the --name argument.

Script What it watches Dedup key
watch_rss.py RSS 2.0 or Atom feed URL <guid> / <id>
watch_http_json.py Any JSON endpoint returning a list of objects Configurable id field
watch_github.py GitHub issues / pulls / releases / commits for a repo id / sha

All three:

  • First run records a baseline — never replays existing feed
  • Watermark is a bounded ID set (max 500) to cap memory
  • Output format: ## <title>\n<url>\n\n<optional body> per item
  • Empty stdout on no-new — the caller treats that as silent
  • Non-zero exit on fetch errors

Usage

Run a watcher directly from the terminal tool:

python $HERMES_HOME/skills/devops/watchers/scripts/watch_rss.py \
  --name hn --url https://news.ycombinator.com/rss --max 5

Watch a GitHub repo (set GITHUB_TOKEN in ~/.hermes/.env to avoid the 60 req/hr anonymous rate limit):

python $HERMES_HOME/skills/devops/watchers/scripts/watch_github.py \
  --name hermes-issues --repo NousResearch/hermes-agent --scope issues

Poll an arbitrary JSON API:

python $HERMES_HOME/skills/devops/watchers/scripts/watch_http_json.py \
  --name api --url https://api.example.com/events \
  --id-field event_id --items-path data.events

Wiring into cron

Ask the agent to schedule a cron job with a prompt like:

Every 15 minutes, run watch_rss.py --name hn --url https://news.ycombinator.com/rss. If it prints anything, summarize the headlines and deliver them. If it prints nothing, stay silent.

The agent invokes the script via the terminal tool inside the cron job's agent loop; no changes to cron's built-in --script flag are needed.

State files

Every watcher writes $HERMES_HOME/watcher-state/<name>.json. Inspect:

cat $HERMES_HOME/watcher-state/hn.json

Force a replay (next run treated as first poll):

rm $HERMES_HOME/watcher-state/hn.json

Writing your own

All three scripts use the same template: load watermark, fetch, diff, save, emit. scripts/_watermark.py is the shared helper; import it to get atomic writes + bounded ID set + first-run baseline for free. See any of the three reference scripts for how little boilerplate it takes.

Common Pitfalls

  1. Printing a "no new items" header every tick. Callers rely on empty stdout = silent. If you print anything on an empty delta, you spam the channel. The shipped scripts handle this; custom scripts must too.
  2. Expecting the first run to emit items. It won't — first run records a baseline. If you need an initial digest, delete the state file after the first run or add a --prime-with-latest N flag in your own script.
  3. Unbounded watermark growth. The shared helper caps at 500 IDs. Raise it for high-churn feeds; lower it on constrained filesystems.
  4. Putting the state dir where the agent's sandbox can't write. $HERMES_HOME/watcher-state/ is always writable. Docker/Modal backends may not see arbitrary host paths.