← All runbooks
gooseworks-ai / composites-newsletter-monitor

Newsletter Monitor

Convert the `newsletter-monitor` skill into a repeatable AgentMail newsletter signal scan. Scan an AgentMail inbox for newsletter signals using configurable keyword campaigns. Extracts matched keywords, context snippets, and company mentions from incoming emails. Use for monitori

agent codexmodel gpt-5.5snapshot python312-uveval programmatic6 stepsv1.0.0

Deploy Newsletter Monitor to your jetty.io

One-click installs this runbook into a collection on your Jetty account. You can run it from the Spot dashboard, schedule it, or pipe inputs in via the API.

Deploy on jetty.io →View source
The shape of the run

6 steps · start to finish.

  1. 1
    Step 1

    Environment Setup

    ▶
    set -euo pipefail

RESULTS_DIR="${RESULTS_DIR:-/app/results}"
mkdir -p "$RESULTS_DIR" "$RESULTS_DIR/work"

if [ -z "${{AGENTMAIL_API_KEY:-}}" ]; then
  echo "ERROR: AGENTMAIL_API_KEY is not set"
  exit 1
fi

command -v python3 >/dev/null || {{ echo "ERROR: python3 not installed"; exit 1; }}
command -v pip3 >/dev/null || {{ echo "ERROR: pip3 not installed"; exit 1; }}

pip3 install agentmail python-dotenv
  2. 2
    Step 2

    Resolve Scan Parameters

    ▶

    Resolve orchestration inputs into CLI arguments. Only include optional flags when values are provided so defaults from the source skill remain intact.

  3. 3
    Step 3

    Run Newsletter Scan

    ▶

    Run both JSON and summary modes. The JSON output is used for downstream processing and validation; the summary output is kept for human review.

  4. 4
    Step 4

    Validate Outputs

    ▶

    Validate that the scan produced parseable JSON and that the required summary artifacts exist.

  5. 5
    Step 5

    Iterate on Errors (max 3 rounds)

    ▶

    If setup, scanning, or output validation fails, retry up to max 3 rounds after applying the relevant correction.

  6. 6
    Step 6

    Write Reports

    ▶

    set -euo pipefail