← All runbooks
gooseworks-ai / capabilities-contact-cache

contact-cache

Use the Contact Cache skill to maintain a CSV-backed database of identified and contacted people across outreach strategies. The workflow prevents duplicate outreach by checking LinkedIn URLs and email addresses before adding or updating records. It also supports status tracking,

agent codexmodel gpt-5.5snapshot python312-uveval programmatic7 stepsv1.0.0

Deploy contact-cache 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.

The shape of the run

7 steps · start to finish.

  1. 1
    Step 1

    Environment Setup

    mkdir -p /app/results
    python3 --version
    if [ ! -f skills/contact-cache/scripts/cache.py ]; then
      echo "ERROR: skills/contact-cache/scripts/cache.py not found"
      exit 1
    fi
    

    Verify that the source skill checkout is available from the current working directory and that /app/results is writable before running cache operations.

  2. 2
    Step 2

    Inspect Inputs

    Normalize any supplied LinkedIn URLs, email lists, status values, and strategy names. If no explicit operation is supplied, run `stats` first so the output files still describe the current cache state.

  3. 3
    Step 3

    Run the Contact Cache Operation

    Use the source CLI exactly as intended by the skill. Common commands are:

  4. 4
    Step 4

    Export Current Cache State

    After any successful mutation, export the cache so downstream systems can inspect the final state.

  5. 5
    Step 5

    Write Structured Results

    Write `/app/results/contact_cache_summary.json` with operation metadata, duplicate counts, added or updated contact IDs when available, export path, and the cache data path. The cache stores contacts in:

  6. 6
    Step 6

    Iterate on Errors (max 3 rounds)

    If setup, execution, export, or validation fails, apply one targeted fix and retry, for max 3 rounds. Common fixes are creating the missing data directory, correcting a status to one of the valid values, quoting comma-separated URL or email arguments, or switching the export form

  7. 7
    Step 7

    Final Checklist

    Run this verification script before finishing: