← Back to Skills Marketplace
Analyst Watchdog
by
Nissan Dookeran
· GitHub ↗
· v1.0.1
· MIT-0
362
Downloads
0
Stars
1
Active Installs
2
Versions
Install in OpenClaw
/install analyst-watchdog
Description
Automated monitoring agent that watches an API scoreboard, detects milestones, writes findings to markdown, and alerts via file-based triggers. Use when you...
README (SKILL.md)
Last used: 2026-03-24 Memory references: 1 Status: Active
Analyst Watchdog
An autonomous monitoring agent pattern: watch a system, detect changes, write findings, alert when thresholds are crossed. Runs on a schedule (LaunchAgent or cron) without human intervention.
Pattern
API Endpoint → Poll → Detect Changes → Write FINDINGS.md
→ Write OUTBOX.md (for orchestrator)
→ Write ALERT_TELEGRAM.md (urgent)
What It Monitors
Configurable, but the reference implementation tracks:
- Model evaluation scores hitting milestones (n=50, 100, 150, 200)
- Promotion events (model proven equivalent to cloud baseline)
- Score anomalies (sudden drops or improvements)
- System health degradation
Alert Tiers
| File | Urgency | Who Reads It |
|---|---|---|
FINDINGS.md |
Low | Background knowledge |
OUTBOX.md |
Medium | Orchestrator on next heartbeat |
ALERT_TELEGRAM.md |
High | Sent immediately, then deleted |
Files
scripts/analyst_agent.py— Reference watchdog implementation
Usage Guidance
This skill appears coherent and limited to local monitoring. Before installing, confirm the control-plane API at localhost:8765 is the intended trusted endpoint and that writing/overwriting files under the skill's workspace (projects/hybrid-control-plane and agents/analyst) is acceptable. If you are concerned about accidental file writes, run the script in a sandbox or inspect/backup the target directories. If you need remote alert delivery, note that this script only writes local alert files; an external orchestrator must pick them up to send messages.
Capability Analysis
Type: OpenClaw Skill
Name: analyst-watchdog
Version: 1.0.1
The analyst-watchdog skill is a standard monitoring utility designed to poll a local API endpoint (localhost:8765) and log performance milestones to local Markdown files. The script (scripts/analyst_agent.py) uses safe standard library functions for network requests and file I/O, with no evidence of data exfiltration, unauthorized execution, or obfuscation. Its behavior aligns perfectly with the documentation provided in SKILL.md.
Capability Assessment
Purpose & Capability
Name/description describe a local monitoring/alerting agent and the implementation requires only python3 and accesses a local control-plane API (http://localhost:8765). Required binaries and behavior are proportionate to the stated purpose.
Instruction Scope
SKILL.md instructs the agent to poll a local API and write FINDINGS/OUTBOX/ALERT files; the included Python does exactly that. It reads project score files and writes state and alert files under the workspace; there are no instructions to read unrelated system secrets or transmit data off-host.
Install Mechanism
No install spec is provided (instruction-only with a reference script). No downloads, package installs, or archive extraction are present.
Credentials
The skill requests no environment variables, credentials, or config paths. The script uses only local filesystem paths relative to the skill workspace and calls localhost endpoints; requested access is minimal and consistent with the functionality.
Persistence & Privilege
always is false (not force-installed). The skill runs as an agent when invoked; it does not modify other skills or system-wide configs. It creates and updates local state and markdown files as expected for an autonomous watchdog.
How to Use
- Make sure OpenClaw is installed (local or Docker)
- Run the install command in chat:
/install analyst-watchdog - After installation, invoke the skill by name or use
/analyst-watchdog - Provide required inputs per the skill's parameter spec and get structured output
Version History
v1.0.1
Add security_notes: local-only monitoring, no external data transmission
v1.0.0
Initial release — autonomous system monitoring agent pattern
Metadata
Frequently Asked Questions
What is Analyst Watchdog?
Automated monitoring agent that watches an API scoreboard, detects milestones, writes findings to markdown, and alerts via file-based triggers. Use when you... It is an AI Agent Skill for Claude Code / OpenClaw, with 362 downloads so far.
How do I install Analyst Watchdog?
Run "/install analyst-watchdog" in the OpenClaw or Claude Code chat to install it in one step — no extra setup required.
Is Analyst Watchdog free?
Yes, Analyst Watchdog is completely free, licensed under MIT-0. You can download, install and use it at no cost.
Which platforms does Analyst Watchdog support?
Analyst Watchdog is cross-platform and runs anywhere OpenClaw / Claude Code is available (cross-platform).
Who created Analyst Watchdog?
It is built and maintained by Nissan Dookeran (@nissan); the current version is v1.0.1.
More Skills