Aoineco Ledger — AI Agent Financial Tracking Engine

Track detailed AI agent expenses, revenues, budgets, and ROI with per-agent attribution and alerts for efficient micro-budget financial management.

Security Scan

The skill mostly behaves like a local ledger (writes a JSONL journal, budgets, CSV export) and asks for no credentials, but some claimed capabilities (encrypted storage / VaultCrypto integration, auto-conversion) and a mismatch between metadata and implementation leave unexplained behavior that should be inspected before trusting it with sensitive data.

Purpose & Capability

The SKILL.md and included Python implement a local ledger for per-agent cost tracking which matches the stated purpose. However the header and docs claim features such as “Encrypted storage via VaultCrypto integration” and “auto-conversion” (multi-currency) while the package declares zero external dependencies and no credentials/config for a vault or FX service — that is an internal inconsistency that should be explained.

Instruction Scope

Runtime instructions in SKILL.md are narrowly scoped: import the AoinecoLedger class and call logging/export methods. The implementation shown only reads/writes files in a ledger directory (default ~/.openclaw/workspace/ledger or OPENCLAW_WORKSPACE override). There are no SKILL.md instructions to read other unrelated files, exfiltrate data, or contact external endpoints.

Install Mechanism

No install spec is provided (instruction-only skill + included script). The code claims 'zero external dependencies' and is pure Python; that is low risk from a supply/install perspective. Still, the codebase includes features that normally require external services (vaults, FX rates) but no install/credential hints for those, which is inconsistent.

Credentials

The skill declares no required environment variables or credentials. The implementation optionally respects OPENCLAW_WORKSPACE for the storage location — a reasonable integration point. There are no requests for API keys, cloud credentials, or unrelated secrets in metadata or visible code.

Persistence & Privilege

The skill is not force-enabled (always=false) and will persist only to its own ledger directory. It creates and writes local files (transactions.jsonl, accounts.json, budgets.json, config.json) under the chosen workspace. It does not request elevation or attempt to modify other skills' configs in the visible code.

What to consider before installing

This package looks like a local bookkeeping library and asks for no secrets, but there are a few red flags to check before installing or using it with real data: - Ask the author or inspect the full ledger_engine.py for any network calls, hard-coded endpoints, or subprocess executions that are not visible in the truncated snippet. The file header promises VaultCrypto encryption and currency auto-conversion — verify how those are implemented and where keys or FX rates come from. - Confirm there is no hidden code that attempts to send the ledger contents to an external server or to read unrelated files on disk. Search the full file for 'requests', 'urllib', 'socket', 'subprocess', 'open', or any HTTP endpoints/URLs. - If you plan to record sensitive billing data, run the skill in an isolated environment (sandbox/container) and review the created ledger directory (~/.openclaw/workspace/ledger by default) to ensure only expected files are written. - If you need encrypted storage, require the implementation detail: how keys are derived/stored and whether the vault integration requires secrets. If the skill claims encryption but offers no key management, do not rely on it. If you can provide the remaining portion of scripts/ledger_engine.py (the truncated part) I can re-evaluate and raise the confidence level or surface specific lines of concern.

<!-- 🌌 Aoineco-Verified | S-DNA: AOI-2026-0213-SDNA-AL01 -->

Version: 1.0.0
Author: Aoineco & Co.
License: MIT
Tags: finance, expense-tracking, budget, roi, bootstrap, multi-agent, accounting

Description

Financial tracking engine built for AI agents operating on extreme budgets. Tracks every API call, gas fee, and revenue stream with per-agent cost attribution. Native support for the $7 Bootstrap Protocol.

Rebuilt from agentledger (Node.js) into Python with major enhancements for multi-agent squads and micro-budget operations.

Core metric: Intelligence per Dollar (IPD) — how many operations you get per cent spent.

Problem

AI agents spend money (API calls, gas fees, subscriptions) but rarely track it. When you're running a 9-agent squad on a $7 seed budget, every fraction of a cent matters. You need:

• Real-time visibility into burn rate and runway
• Per-agent cost attribution (who's spending what?)
• Budget alerts BEFORE you overspend
• Revenue tracking to measure actual ROI

Features

Feature	Description
Transaction Logging	Log expenses, revenue, API costs, gas fees with one-liner methods
$7 Bootstrap Metrics	Seed tracking, ROI%, runway days, daily burn rate, IPD
Per-Agent Cost Attribution	Track spending by squad member (oracle, blue-sound, etc.)
Budget Alerts	Set limits per category/period — warns at 80%, blocks at 100%
Category System	Pre-configured for AI ops: API/LLM, Gas/Blockchain, Revenue/Gig, etc.
CSV/JSON Export	Full transaction export for tax, audit, or analysis
JSONL Storage	Append-only transaction log — fast, corruption-resistant
Period Filtering	Query by today, this-week, this-month, this-year, or all

Quick Start

from ledger_engine import AoinecoLedger

ledger = AoinecoLedger()

# Log API cost
ledger.log_api_cost(0.0042, "Google", "gemini-3-flash", tokens_used=150000, agent="oracle")

# Log gas fee
ledger.log_gas(0.0003, chain="base", tx_hash="0xabc123", agent="blue-sound")

# Log revenue
ledger.log_revenue(0.01, "MoltLaunch", "Tier-1 Intel Report", category="Revenue/Gig")

# Set budget
ledger.set_budget("API/LLM", limit=3.00, period="daily")

# Check $7 Bootstrap metrics
metrics = ledger.get_bootstrap_metrics()
# → Seed: $7.00, Burn: $0.02/day, Runway: 300 days, IPD: 257.5 ops/$

Bootstrap Metrics Output

💰 Seed Amount: $7.00
📈 Total Revenue: $0.0150
📉 Total Expenses: $0.0233
🤖 API/LLM Cost: $0.0230
💵 Net Profit: $-0.0083
📊 ROI: -0.1%
💎 Remaining Balance: $6.9917
🔥 Daily Burn Rate: $0.0233/day
⏱️ Runway: 300 days
🧠 IPD: 257.5 ops/$

Pre-Configured Categories

• API/LLM — OpenAI, Anthropic, Google API costs
• Gas/Blockchain — On-chain transaction fees
• Revenue/Gig — MoltLaunch gig income
• Revenue/DeFi — Meteora, LP yields
• Revenue/Music — claw.fm royalties
• Infrastructure, Marketing, Tools, Subscriptions, Other

File Structure

aoineco-ledger/
├── SKILL.md              # This file
└── scripts/
    └── ledger_engine.py  # Main engine (zero external dependencies)

Zero Dependencies

Pure Python 3.10+. No pip install needed. Designed for the $7 Bootstrap Protocol — every byte counts.