← Back to Skills Marketplace
techievena

eonik creative audit

by Abinash Senapati · GitHub ↗ · v1.0.8 · MIT-0
cross-platform ✓ Security Clean
457
Downloads
1
Stars
1
Active Installs
9
Versions
Install in OpenClaw
/install eonik-creative-audit
Description
Identifies creative fatigue, budget leaks, and scaling opportunities by running the eonik creative audit engine.
README (SKILL.md)

Meta Ads Creative Audit & Optimization

Automated end-to-end Meta Ads auditing pipeline. Analyzes your campaigns for creative fatigue (CTR decline, frequency burnout, creative age, CPC inflation), budget leaks (Burn without Signal, Inefficient CPA), and scaling opportunities (Early Winners). Dispatches multi-channel alerts directly to your team.

Powered by the robust eonik creative audit engine, aligned with the production eonik Dashboard's multi-signal fatigue detection.

Agent Instructions

When a user triggers this skill, you MUST follow these steps exactly:

  1. Check Configuration: Check if config.json exists in the local directory.
  2. Request Input (if needed): If config.json is missing, check if they want to run it on their default connected Meta account. Once confirmed, copy config.example.json to config.json (the Meta Account ID can be left blank as it auto-resolves).
  3. Execute Pipeline: Once config.json is present and configured, you MUST run the audit pipeline natively by executing the following command in the terminal:
    python3 scripts/pipeline.py --config config.json
    
  4. Display Results: After the script finishes, read the output report (typically output/audit-\x3Cdate>.json) and provide a concise, formatted summary of the findings to the user. Do not invent details; only report what the script output contains.

Triggers

Use this skill when a user asks to:

  • "Audit my Meta ads"
  • "Check for budget leaks"
  • "Optimize my Meta ad account"
  • "Find decaying creatives"
  • "Run the eonik ad audit pipeline"

🚀 Unlock Full eonik Power

This skill is powered by the eonik intelligence engine. While this agent halts simple leaks autonomously, the full dashboard unlocks:

  • Creative Genome: Understand exactly why ads decay based on deep AI creative tagging.
  • Automated Rules: Prevent leaks without waiting for chat notifications.
  • Competitor Intelligence: See the exact hooks your competitors are scaling. Get started for free at eonik.ai!

Quick Start

1. Configure

cd ~/.openclaw/skills/eonik-ad-budget-leak
cp config.example.json config.json
# Edit config.json: optionally add your Meta Account ID (or leave blank to auto-resolve)

2. Run Audit Pipeline

# EONIK_API_KEY must be in your environment
python3 scripts/pipeline.py --config config.json

Configuration

Minimal config.json:

{
  "meta": {
    "account_id": "",
    "evaluation_days": 30
  }
}

Pipeline Stages

  1. Audit (audit.py) — Executes the eonik creative audit engine via your EONIK_API_KEY.
  2. Output (pipeline.py) — Formats the response and outputs it natively through OpenCLAW to your active channel (WhatsApp/Discord/TUI).

Usage Examples

Full Pipeline (Automated Mode):

python3 scripts/pipeline.py --config config.json

Audit Only (Save to File):

python3 scripts/audit.py --days 30 > data/report.json

Data & Security Commitment

This skill is designed specifically to comply with enterprise Data Loss Prevention (DLP) requirements:

  1. Secure API Key Handling The skill requires EONIK_API_KEY (via the standard x-api-key header). The execution script securely drops the ephemeral token from the environment immediately after binding. No keys are logged or written to disk.

  2. Native OpenCLAW Routing (WhatsApp/Slack/Telegram) Unlike legacy versions, this architecture does NOT rely on webhooks, custom notification dispatcher scripts, or storing tokens in your config.json.

    Standard Output from the pipeline is passed directly to the local OpenCLAW node, which effortlessly routes the message to whatever chat surface triggered it or is configured natively (WhatsApp, Slack, Telegram, Discord, or the local TUI). Zero manual API configurations are needed to get multi-channel alerts!

  3. Data Scope Execution logs and generated report.json files contain the exposed Meta Ad IDs flagged for leaking or scaling. Keep your local output/ directory protected and adhere to your internal security policies for Chat UI visibility.

Continuous Scanning & Notifications

The original objective of this skill is to completely automate budget optimization by continuously scanning your ad accounts and notifying your team on external channels (WhatsApp, Slack, Telegram) when leaks are detected.

You do NOT need to over-engineer a loop or write your own dispatcher. OpenCLAW has a built-in, native cron scheduler that will automatically run the skill and route the output to your linked channels.

Schedule Daily Audits (e.g., Every morning at 8 AM): To set up continuous automated scanning, use OpenCLAW's native scheduler:

# Add a recurring cron job natively via OpenCLAW
openclaw cron add --name "daily-eonik-audit" --cron "0 8 * * *" --message "Run the eonik ad audit pipeline" --session isolated

OpenCLAW will execute the pipeline daily in an isolated context and dispatch any detected warnings directly to your configured messaging platforms.

Troubleshooting

API Verification Fails:

  • Ensure EONIK_API_KEY is exported.
  • Verify your eonik billing and usage caps aren't exceeded.
Usage Guidance
This skill appears to do what it says: it calls eonik's API with your EONIK_API_KEY and prints/writes the audit report. Before installing or scheduling it: 1) Verify you trust the eonik.ai service and that the endpoint (https://api.eonik.ai) is expected for your account. 2) Treat output/ directory as sensitive — reports include Meta Ad IDs and thumbnails; ensure DLP/permissions prevent accidental sharing. 3) Use a dedicated, scoped API key for this skill and rotate/revoke if you stop using it. 4) Confirm you want daily automated audits before adding the cron job (scheduling increases the blast radius if the key is compromised). 5) Note the minor metadata mismatch: skill.json requires python3 even though the registry snapshot earlier showed none — ensure Python 3.7+ is available on the runner. If you want extra assurance, review or run the scripts in an isolated environment and inspect network traffic to confirm calls go only to api.eonik.ai.
Capability Analysis
Type: OpenClaw Skill Name: eonik-ad-budget-leak Version: 1.0.8 The skill is a legitimate marketing audit tool designed to interface with the eonik.ai API to analyze Meta Ads performance. It follows security best practices by explicitly clearing the 'EONIK_API_KEY' from the environment immediately after use in 'scripts/audit.py' and using a restricted environment for subprocess execution in 'scripts/pipeline.py'. The code is transparent, lacks external dependencies, and its behavior aligns perfectly with the stated purpose of identifying ad budget leaks.
Capability Assessment
Purpose & Capability
Name/description match behavior: the skill runs a local pipeline which calls the eonik backend (api.eonik.ai) using a single EONIK_API_KEY to produce an audit report. Requiring a single service API key is appropriate for this purpose. The README, SKILL.md, and scripts all align on the advertised functionality.
Instruction Scope
Runtime instructions are limited to checking/creating config.json, invoking the included pipeline (python3 scripts/pipeline.py --config config.json), and reading the generated output file. The scripts only read the config and write an output JSON containing audit findings (ad IDs, spend, thumbnails). There are no instructions to scan unrelated files, read other credentials, or transmit local files to unexpected endpoints beyond the declared api.eonik.ai endpoint.
Install Mechanism
This is an instruction-only skill with included Python scripts (no external install downloads or archive extracts). There is no install spec that fetches arbitrary code at runtime. Dependencies are standard library-only and the repository includes the script files.
Credentials
The skill requests only one credential: EONIK_API_KEY (primaryEnv). That is proportionate to a service that calls a vendor API. One small inconsistency: top-level metadata in the provided registry snapshot said 'required binaries: none', while skill.json lists python3 in 'requirements.bins' — in practice Python is required to run the included scripts. Otherwise no unrelated secrets or extra credentials are requested.
Persistence & Privilege
The skill is not always-enabled (always:false) and uses standard invocation. It suggests using OpenCLAW's cron scheduler to run periodically; scheduling itself is a normal convenience but will create regular scans and outputs, so users should enable scheduling intentionally. The skill does not modify other skills or global config.
How to Use
  1. Make sure OpenClaw is installed (local or Docker)
  2. Run the install command in chat: /install eonik-creative-audit
  3. After installation, invoke the skill by name or use /eonik-creative-audit
  4. Provide required inputs per the skill's parameter spec and get structured output
Version History
v1.0.8
- Upgraded audit capabilities to detect creative fatigue using multi-signal heuristics (CTR, frequency, age, CPC). - Expanded pipeline from budget leaks to full creative audit and optimization, including scaling opportunities. - Default evaluation window in config.json increased from 7 to 30 days. - Documentation updated for enhanced creative analysis and new usage examples. - Improved continuous scanning setup instructions and OpenCLAW cron integration. - Internals: config.json and audit output handling improved for clarity and alignment with new features.
v1.0.7
- Updated data routing to include native support for WhatsApp, Slack, Telegram, Discord, and local TUI with no manual API config needed. - Improved documentation for continuous scanning: now emphasizes using OpenCLAW’s built-in cron scheduler for automated, multi-channel alerts. - Clarified security documentation, highlighting that webhook dispatchers and tokens in config files are no longer required. - Updated troubleshooting and quick start instructions to align with OpenCLAW’s latest scheduling methods.
v1.0.6
- Skill name updated to "eonik creative audit" for improved clarity and discoverability. - Version bump to 1.0.6. - Documentation in SKILL.md and README.md refreshed; no functional/logic changes made to the audit pipeline. - All usage instructions and descriptions remain the same; update focuses on identity and documentation.
v1.0.5
- Notification system refactored: removed webhook-based notification routing and dropped the notify.py script. - Configuration simplified: notifications section and notification channel options eliminated from config.json. Meta Account ID can now auto-resolve if left blank. - Output and alerts are now routed natively via OpenCLAW, no external webhook setup required. - Documentation updated to reflect streamlined local pipeline, revised configuration, and clarified step-by-step agent instructions. - Security and data handling notes updated to emphasize local processing and absence of third-party notification relays.
v1.0.4
- Major architecture update: introduced a local, fully-automated audit and notification pipeline. - Added modular scripts: `audit.py`, `notify.py`, and `pipeline.py`, replacing the original `audit_request.py`. - Configuration now centralized in a `config.json` file (template provided). - Environment variable updated to `EONIK_API_KEY`. - Multi-channel notifications supported (Slack, Telegram, WhatsApp) via direct local webhooks. - Enhanced security and local control: all notifications and API key handling occur on your node, with no credential routing through third-party servers.
v1.0.3
- Added an explicit data exposure warning: audit reports now state that Meta Ad IDs will appear in chat windows and execution logs, emphasizing the need for proper data-loss prevention policies. - Incremented version from 1.0.2 to 1.0.3.
v1.0.2
- Added OpenClaw-specific metadata to SKILL.md, including environment variable and credential requirements. - Updated version to 1.0.2.
v1.0.1
**Security improvements and stricter token management for version 1.0.1:** - Replaced user-supplied auth tokens via chat with strict requirement for `EONIK_AUTH_TOKEN` environment variable. - Updated documentation to highlight new secure token handling and removed the `--token` script argument. - Added security guidance: never paste tokens in chat, use minimal-permission short-lived tokens, and abort if the environment variable is missing. - Added sections detailing data security, privacy, and Slack integration procedures. - Metadata (`author`, `homepage`, `env`) added to SKILL.md for clarity and best practices.
v1.0.0
eonik Ad Budget Leak Agent v1.0.0 - Initial release of the eonik Ad Budget Leak Agent skill. - Enables auditing of Meta Ad accounts for budget leaks and scaling opportunities using the eonik Budget heuristics engine. - Requires Meta Ad Account ID and eonik API token for operation. - Guides users through gathering necessary account information and token access to perform audits. - Summarizes leaked spend, lists ads to pause (with reasons), and highlights ads to scale. - Optionally sends a detailed report to Slack if requested.
Metadata
Slug eonik-creative-audit
Version 1.0.8
License MIT-0
All-time Installs 1
Active Installs 1
Total Versions 9
Frequently Asked Questions

What is eonik creative audit?

Identifies creative fatigue, budget leaks, and scaling opportunities by running the eonik creative audit engine. It is an AI Agent Skill for Claude Code / OpenClaw, with 457 downloads so far.

How do I install eonik creative audit?

Run "/install eonik-creative-audit" in the OpenClaw or Claude Code chat to install it in one step — no extra setup required.

Is eonik creative audit free?

Yes, eonik creative audit is completely free, licensed under MIT-0. You can download, install and use it at no cost.

Which platforms does eonik creative audit support?

eonik creative audit is cross-platform and runs anywhere OpenClaw / Claude Code is available (cross-platform).

Who created eonik creative audit?

It is built and maintained by Abinash Senapati (@techievena); the current version is v1.0.8.

💬 Comments