← Back to Skills Marketplace
Data Anomaly Detector
by
datadrivenconstruction
· GitHub ↗
· v2.1.0
4558
Downloads
1
Stars
39
Active Installs
2
Versions
Install in OpenClaw
/install data-anomaly-detector
Description
Detect anomalies and outliers in construction data: unusual costs, schedule variances, productivity spikes. Statistical and ML-based detection methods.
Usage Guidance
Install only if you are comfortable granting filesystem access for construction or business data you choose to analyze. Provide specific datasets, review generated findings before relying on them, and confirm export paths before allowing reports to be written.
Capability Analysis
Type: OpenClaw Skill
Name: data-anomaly-detector
Version: 2.1.0
The skill bundle is designed for data anomaly detection in construction data. The Python code in SKILL.md uses standard data science libraries (pandas, numpy, scipy) for statistical analysis and report generation, without any network calls or shell command execution. The `claw.json` explicitly declares 'filesystem' permission, which is consistent with the `pd.read_excel` usage shown in the quick start guide for loading user-provided data. Neither the markdown instructions nor the code exhibit any signs of malicious intent, such as data exfiltration, persistence mechanisms, or prompt injection attempts to subvert the agent's purpose. All components align with the stated benign functionality.
Capability Assessment
Purpose & Capability
The manifest, instructions, and SKILL.md consistently describe detecting anomalies in construction costs, schedules, productivity, duplicates, sequence gaps, and data quality issues.
Instruction Scope
The instructions mention creating cost estimates and offering exports, which is broader than pure anomaly detection, but it is disclosed in the construction cost-management context and does not add account access, mutation authority, or hidden execution.
Install Mechanism
The bundle contains JSON/Markdown instructions and sample Python code only; there are no install hooks, executables, remote installers, or automatic scripts.
Credentials
Filesystem permission is proportionate for reading user-supplied CSV, Excel, JSON, or direct input and for optional exports, but users should be deliberate about input and output paths.
Persistence & Privilege
No credentials, network endpoints, background workers, durable stores, shell execution, deletion behavior, or privilege escalation are present; the sample history is in-memory only.
How to Use
- Make sure OpenClaw is installed (local or Docker)
- Run the install command in chat:
/install data-anomaly-detector - After installation, invoke the skill by name or use
/data-anomaly-detector - Provide required inputs per the skill's parameter spec and get structured output
Version History
v2.1.0
- Added detailed documentation and usage examples to SKILL.md for construction anomaly detection.
- Clarified support for statistical (IQR, z-score) and business rule-based anomaly detection for costs and schedules.
- Described construction-specific thresholds and anomaly types handled (cost, schedule, productivity).
- Provided technical overview and sample Python implementation in SKILL.md.
v1.0.0
Data Anomaly Detector 1.0.0 – Initial Release
- Detects anomalies and outliers in construction data, including unusual costs, schedule variances, and productivity spikes.
- Utilizes statistical (IQR, z-score) and business rule-based techniques for anomaly detection.
- Identifies cost overruns, negative costs, and group-specific outliers.
- Flags schedule data issues like negative or excessive task durations.
- Supports automatic anomaly reporting with severity, context, and suggested actions.
Metadata
Frequently Asked Questions
What is Data Anomaly Detector?
Detect anomalies and outliers in construction data: unusual costs, schedule variances, productivity spikes. Statistical and ML-based detection methods. It is an AI Agent Skill for Claude Code / OpenClaw, with 4558 downloads so far.
How do I install Data Anomaly Detector?
Run "/install data-anomaly-detector" in the OpenClaw or Claude Code chat to install it in one step — no extra setup required.
Is Data Anomaly Detector free?
Yes, Data Anomaly Detector is completely free (open-source). You can download, install and use it at no cost.
Which platforms does Data Anomaly Detector support?
Data Anomaly Detector is cross-platform and runs anywhere OpenClaw / Claude Code is available (darwin, linux, win32).
Who created Data Anomaly Detector?
It is built and maintained by datadrivenconstruction (@datadrivenconstruction); the current version is v2.1.0.
More Skills