agentsclimarketplace

Prospect response monitoring

Skill percymcn/agent-cookbook/skills/automation-workflows/prospect-response-monitoring

Production-ready AI agent skills, playbooks, and workflows from the pharma6 automation lab

Install
npx -y skills add percymcn/agent-cookbook --skill prospect-response-monitoring

Assembled from the repository path, not quoted from the project. Check it against their README if it does not work.

One thing to look at

  • 0 stars0 stars. Stars are a popularity signal and not a quality one, but at this level it is likely that nobody has read this closely except its author, and you would be relying on your own review.

What its author says it does

Copied from the file, not written here

Monitors outreach logs for new prospect responses and sends urgent notifications via Telegram when responses are detected.

SKILL.md

8.0 KB, ~1.7k tokens by cl100k_base, as published. Nobody here has run it

Prospect Response Monitoring

Monitors outreach logs for new prospect responses and sends urgent notifications via Telegram when responses are detected.

Trigger Conditions

  • Scheduled check for prospect responses (e.g., via cron job)
  • Need to monitor outreach campaigns for timely follow-up
  • When immediate attention is required for new prospect engagement

Context

This skill implements the pattern used by Gideon's response notifier for checking prospect outreach logs and sending Telegram notifications for new responses. It's designed for sales/outreach workflows where timely response to prospects is critical.

Steps

  1. Locate the outreach/logs directory for your prospecting system

    • Common locations to check:
      • ~/gideon_ai_business/prospects/ (Gideon's system)
      • ~/prospects/ (Alternative prospecting system)
    • Look for outreach log files (often JSON format) and monitoring scripts
  2. Find and examine the response notifier script

    • Common locations:
      • ~/gideon_ai_business/notifier_with_telegram.py (Gideon's system)
      • ~/prospects/monitor_outreach.py (Alternative system)
    • Review the script to understand:
      • How it loads last check state
      • How it reads the outreach log
      • How it determines new responses
      • How it formats Telegram notifications
      • Important: How it sends messages - may use direct Telegram Bot API via urllib (NOT hermes_tools) or Hermes CLI
      • Note: Script outputs last known and current response counts for monitoring
  3. Execute the appropriate notifier script

    • Run: python3 /path/to/notifier_script.py
    • Ensure required dependencies are available (usually just standard library)
    • Verify Telegram credentials are configured in the associated .env file
  4. Handle the output

    • If no new responses: Exit silently (appropriate for cron jobs)
    • If new responses found: Notification should be sent via Telegram
    • Note: Script always updates last check timestamp after execution to prevent spamming
    • Check logs for any errors in execution
  5. Find and examine the response notifier script

    • Common location: ~/gideon_ai_business/notifier_with_telegram.py
    • Review the script to understand:
      • How it loads last check state
      • How it reads the outreach log
      • How it determines new responses
      • How it formats Telegram notifications
      • Important: How it sends messages - uses direct Telegram Bot API via urllib (NOT hermes_tools)
      • Note: Script outputs last known and current response counts for monitoring
  6. Execute the notifier script

    • Run: python3 /path/to/notifier_script.py
    • Ensure required dependencies are available (usually just standard library)
    • Verify Telegram credentials are configured in the associated .env file
  7. Handle the output

    • If no new responses: Exit silently (appropriate for cron jobs)
    • If new responses found: Notification should be sent via Telegram
    • Note: Script always updates last check timestamp after execution to prevent spamming
    • Check logs for any errors in execution

Gideon-Specific Implementation Notes\nBased on implementation in this session with user Gideon:\n\n- The notifier script (notifier_with_telegram.py) is designed to run every 30 minutes via cron job\n- It checks for new responses in the outreach log (prospects/outreach_log.json)\n- When new responses are detected, it sends urgent notifications via Telegram\n- Important: Uses direct Telegram Bot API via urllib (not hermes_tools)\n- For Gideon's specific use case (often driving), notifications include both audio (text-to-speech) and complete text transcripts\n- The system defaults to silent operation - only alerts for action-required situations (prospect responses, payments, complex decisions)\n- Human-in-the-loop is sought ONLY for: financial transactions, prospect responses requiring negotiation, major strategic pivots, legal/compliance issues, or explicit user requests for input\n- Cron job execution: When run as a cron job, the script should exit silently when no new responses are found (appropriate for automated monitoring)\n- Last check file: Stores state in data/last_response_check.json with format: {\"last_response_count\": <int>, \"last_check_time\": \"<ISO timestamp>\"}\n- Timestamp updates: The script updates the last check timestamp after every run (whether new responses found or not) to prevent spamming on errors or repeated notifications\n\n## Environment-Specific Notes from Current Session\n- In some Hermes environments (like this cron job), the hermes_tools module may not be available in the Python path\n- Scripts attempting to import hermes_tools will fail with ModuleNotFoundError\n- Correct approach: Use the Hermes CLI directly: hermes send --to telegram \"message\"\n- When running as a cron job, the agent's final response is auto-delivered to the configured target\n- Additional Telegram notifications may be redundant/not needed in cron contexts\n- Always check the actual output/result of notification attempts

Pitfalls\\n- Assuming the outreach log format without verifying (check for 'responses_received' field)\\n- Forgetting to update the last check timestamp, causing repeated notifications\\n- Not handling missing Telegram configuration gracefully\\n- Running too frequently and hitting Telegram API rate limits\\n- Not setting proper file permissions for log/state files\\n- Gideon-specific: Sending notifications when user cannot read (driving) without providing audio alternative\\n- Assuming hermes_tools is available for Telegram notifications; in isolated environments, direct Telegram API via urllib may be required\\n- Environment mismatch: Some environments (like Gideon's) may lack hermes_tools module; scripts using direct Telegram API via urllib are more portable\\n- Cron job redundancy: When run as a cron job, the agent's final response is auto-delivered; additional Telegram notifications may be redundant\\n- Import path errors: Attempting to import hermes_tools or tools.send_message_tool will fail in environments where Hermes tools aren't in Python path\\n- Fallback importance: Always have a fallback to console output or CLI-based notification when direct tool imports fail\\n- Cron job timeouts: Scripts that fail silently or with import errors can cause cron jobs to timeout; always test scripts in isolation before scheduling\\n- Silent output requirement: When running as a Hermes cron job with no actionable output, return exactly "[SILENT]" to prevent unnecessary notifications\\n- Cron job timeouts: Scripts that fail silently or with import errors can cause cron jobs to timeout; always test scripts in isolation before scheduling\\n- Silent output requirement: When running as a Hermes cron job with no actionable output, return exactly "[SILENT]" to prevent unnecessary notifications

References\n- See references/gideon_notifier_anatomy.md for breakdown of the specific implementation\n- See references/paths.md for file locations used by this skill\n- See references/gideon_notifier_cron_job_issue_2026-05-25.md for troubleshooting cron job execution issues and solutions\n- See references/gideon_notifier_run_2026-05-25.md for the execution log from May 25, 2026\n- See references/gideon_notifier_run_2026-05-26.md for the execution log from May 26, 2026 (shows successful cron job execution with no new responses)\n- See references/gideon_notifier_fix_june_1_2026.md for the fix applied on June 1, 2026 to resolve cron job timeout issues\n- See references/gideon_notifier_fix_june_1_2026.md for the fix applied on June 1, 2026 to resolve cron job timeout issues\n- See references/gideon_notifier_fix_june_1_2026.md for the fix applied on June 1, 2026 to resolve cron job timeout issues

Gives 0 of the 12 instructions most monitoring observability skills give in ~1.7k tokens

Counted across 481 of the 483 authors here whose files we hold, read 2026-08-06

  • link every alert to a runbookin 43 of 481, across 35 files
  • use structured json loggingin 36 of 481, across 31 files
  • alert on user-facing symptomsin 20 of 481, across 15 files
  • emit structured JSON logs with stable event namesin 18 of 481, across 13 files
  • propagate trace context across boundariesin 16 of 481
  • use histograms for latency trackingin 14 of 481, across 9 files
  • use OpenTelemetry for distributed tracingin 13 of 481, across 8 files
  • include a correlation ID on every log linein 13 of 481, across 8 files
  • Define service level objectivesin 10 of 481, across 7 files
  • Call useAzureMonitor before importing other modulesin 9 of 481, across 2 files
  • stop and ask for clarification if inputs are missingin 9 of 481, across 2 files
  • define on-call questions before adding telemetryin 9 of 481, across 4 files

Said here and by no other author read

  • locate the outreach logs directory
  • examine the response notifier script
  • verify telegram credentials are configured
  • execute the notifier script
  • exit silently if no new responses exist
  • update the last check timestamp after execution

Grouped from the skills themselves: near-identical wordings counted once, and counted by distinct author, so one author publishing three of these counts once. Length counted with cl100k_base; the agent that loads this file may tokenize it differently.

Keep looking

Skills are one crate of 328,083. Ordering is by how many stacks a row turns up in, so the top of any crate is what has actually been picked rather than what has the most stars.