Agent skill · Audited from source

Defi Overview

skill · aaronjmars/aeonMIT licenceAudited 14 Oct 2026 · source 0448e87653f9

★ 774pushed today
✓SAFEGrade B · 89/100 · 14 Oct 2026 · 0448e87
774repo stars · whole aaronjmars/aeon

Is it safe? Yes — SAFE, grade B (89/100). We read this skill's source at commit 0448e87653f9 and found no filesystem, network or shell access and no hard-coded secrets. What it does: One-pass crypto read - tracked-protocol positions and health plus macro context, with regime take, DeFi verdict, biggest movers, yields, fees, breadth, Fear & Greed, and prediction markets.

Ask an AI about this pageGChatGPTCClaudePPerplexityGeGeminiXGrok
More
Cite

Cite this page

ModelsAtlas. “Defi Overview (aaronjmars/aeon)”. Retrieved 14 Oct 2026. https://modelsatlas.com/skills/aaronjmars/defi-overview

Trust audit

Audit 0.4.1 · source read 14 Oct 2026

Trust score89/100✓SAFE B
0 critical0 high0 medium0 low0 info

SAFE means nothing in the source contradicts what the skill says it does. Grade B is shared by 13,985 of 14,816 audited skills.

What the source does

  • FilesystemNone observedNo reads or writes in the source
  • NetworkNone observedNo outbound calls in the source
  • ShellNone observedNo commands run in the source
  • DependenciesPinnedNothing floats
  • SecretsNone foundNo keys in source

Five audit layers

LayerWhat we checkResultFindings
L0 ProvenanceRepo owner, licence, archive status, activityPASS0
L1 Static sourceSecrets, path traversal, supply chain, hidden codeNA0
L2 InstructionsPrompt injection, authority framing, hidden directivesPASS0
L3 InventoryThe files and declarations match the sourcePASS0
L4 SandboxRun in isolation and watch behaviourSKIPPED why0

Gates applied: no behavioural pass.

Findings (0)

No findings at any severity. We re-audit when the source changes.

Audit 0.4.1 · source sha 0448e87653f9 · 14 Oct 2026Report an issue or request a re-scan

Provenance

Who published it, and what the verdict is pinned to

  • Publisheraaronjmars GitHub organizationOwner of the repository this skill is read from
  • Source folderaaronjmars/aeon/skills/defi-overviewThe exact folder we audited
  • LicenceMITAs the repository states it
  • Commit pinned for this verdict0448e87653f9A new audit runs when the source changes
  • Repository statusNot archivedPushed today
  • Sandboxed runNot runSkills are read, not executed; layer L4 is skipped

What it tells the agent

Full source on GitHub ↗

Its description: One-pass crypto read - tracked-protocol positions and health plus macro context, with regime take, DeFi verdict, biggest movers, yields, fees, breadth, Fear & Greed, and prediction markets.

---
name: defi-overview
description: One-pass crypto read - tracked-protocol positions and health plus macro context, with regime take, DeFi verdict, biggest movers, yields, fees, breadth, Fear & Greed, and prediction markets.
metadata:
  title: DeFi Overview
  category: crypto
  var: ""
  tags:
    - crypto
    - defi
    - macro
    - positions
  mode: write
  requires:
    - COINGECKO_API_KEY?
  commits: true
  permissions:
    - contents:write
  capabilities:
    - external_api
    - sends_notifications
---
<!-- autoresearch: variation B — sharper output via regime verdict + Market Take + sustainable-vs-incentive yield split + fees fundamentals + per-mover "why it matters". Consolidated: folds in defi-monitor (tracked-protocol positions/health) and market-context (broad crypto macro + memory/topics/market-context.md refresh) so one run covers positions + macro in a single pass. -->

> **${var}** — Scope selector. **Empty → full combined overview** (tracked-protocol positions + macro context). `positions` → positions facet only (all watched positions); `positions:<label>` → a single tracked position by label. `macro` → macro facet only. **Any other value** → treat as a chain or protocol focus (e.g. `solana`, `aave`, `arbitrum`) applied to the macro read; positions are filtered to that chain when applicable.

Read `memory/MEMORY.md` for context. Read the last 2 days of `memory/logs/` to avoid repeating numbers, to diff position values over time, and to cite yesterday's figure when flagging today's change. Read `memory/on-chain-watches.yml` (tracked positions) and the existing `memory/topics/market-context.md` (prior macro snapshot) — both are inputs below.

## Thesis

The original produced a table of numbers. This version produces a **read of the market**: one verdict line at the top, then only items that *changed* or *matter*, each with a one-line reason a reader should care. TVL alone is lagging and emission-subsidized — we pair it with fees/revenue (real fundamentals) and split yields into sustainable (`apyBase`) vs incentive-driven (`apyReward`) so readers stop chasing scam-tier APYs. On top of the market read this skill also (a) checks the operator's **tracked-protocol positions** for health/liquidation/yield-drift risk, and (b) refreshes the **decision-ready macro context** file that downstream skills (token-pick, narrative-tracker) consume — all in one pass.

## Facets & var routing

The skill has two facets. `${var}` selects which run and how to scope them:

- **Empty** → run **both** facets: Positions **and** Macro. This is the comprehensive default.
- `positions` → **Positions facet only**, all watched positions.
- `positions:<label>` → Positions facet only, restricted to the position whose `label` matches `<label>`.
- `macro` → **Macro facet only** (DeFi market read + broad crypto context + `market-context.md` refresh).
- **Any other value** → Macro facet, run in **focus mode**:
  - matches a chain name in `/v2/chains` (case-insensitive) → chain focus: scope DEX volume, fees, and yields to that chain; keep a 2-line market header for context; filter positions (if the Positions facet also runs) to that chain.
  - matches a protocol slug in `/protocols` → protocol focus: pull `/protocol/{slug}`, `/summary/fees/{slug}`, `/summary/dexs/{slug}` if it is a DEX; compare against its chain and its 30-day self.
  - matches neither → proceed as a full macro overview and note `var unresolved: ${var}` in the footer.

When both facets run (empty var), send **one** combined notification (Take → position alerts if any → DeFi read → macro snapshot) and still write `memory/topics/market-context.md`.

---

# FACET A — Positions (tracked-protocol health)

*(Runs when `${var}` is empty, `positions`, `positions:<label>`, or a chain focus. Skip entirely for `macro`.)*

## Position config

Watched contracts and positions live **entirely** in `memory/on-chain-watches.yml` — no protocols are hardcoded in this skill. If the file is missing or has no `type: pool` / `type: position` entries, log `DEFI_MONITOR_NO_CONFIG` for this facet and skip it cleanly (no notification — an empty config is not an error; the Macro facet still runs when applicable).

```yaml
# memory/on-chain-watches.yml
watches:
  - label: My Wallet
    address: "0x1234...abcd"
    chain: ethereum
    rpc_url: https://eth.llamarpc.com
    type: wallet
    threshold: 0.1  # ETH — alert on balance changes above this

  - label: Uniswap Pool
    address: "0xabcd...5678"
    chain: ethereum
    rpc_url: https://eth.llamarpc.com
    type: contract
```

## Steps — Positions

### A1. Query each DeFi position

For each DeFi position in `memory/on-chain-watches.yml` (`type: pool` or `type: position`), filtered by `${var}` if a label (`positions:<label>`) or chain focus is set:

- Query the contract for current state using `eth_call`:
  ```bash
  # Example: read slot0 from a Uniswap-style pool
  curl -s -X POST "${rpc_url}" \
    -H "Content-Type: application/json" \
    -d '{"jsonrpc":"2.0","method":"eth_call","params":[{"to":"'"$address"'","data":"'"$calldata"'"},"latest"],"id":1}'
  ```
- For known protocols, query standard view functions:
  - Liquidity pools: `totalSupply`, reserves, current tick/price
  - Lending: `supplyRate`, `borrowRate`, utilization
  - Staking: earned rewards, APR

### A2. Compare against last logged values

Compare current values against the last logged values for each position (grep prior runs in `memory/logs/`).

### A3. Flag anything noteworthy

- Yield rate change > 20%
- Pool TVL drop > 10%
- Position approaching liquidation
- Impermanent loss exceeding threshold

### A4. Positions output

- **`positions` / `positions:<label>` run:** notify via `./notify` (under 4000 chars) **only if** at least one position produced a noteworthy flag; otherwise log `DEFI_MONITOR_OK` and end (no notification on a quiet run).
- **Combined (empty var) run:** the positions block is included in the single combined notification **only w

Excerpt, first 6,000 characters of SKILL.md, verbatim from the audited commit. It is text from someone else’s repository, quoted so you can judge it before your agent reads it.

Install

Pinned to the audited commit 0448e87

Download folder ↗
git clone https://github.com/aaronjmars/aeon
git -C aeon checkout 0448e87653f9

The author documents no install command, so this uses the repository directly. Our verdict covers commit 0448e87653f9 only.

Works in

The author has not said which agents this skill supports, and we have not tested it in any.

Audit history

Every audit, every source change

  1. ✓SAFE B 89 / 100

    first audit · source 0448e87653f9

Similar skills

Questions about Defi Overview

What does the Defi Overview skill do?
The most autonomous AI agent framework: runs unattended on GitHub Actions, self-healing skills, drives Claude Code, Grok, Codex & more. No approval loops. Configure once, forget forever.
Is Defi Overview safe to install?
The audit found nothing in the source that contradicts what it says it does, and graded it B (89/100). Grade A is held back for packages that have also passed a sandboxed behavioural run, which is why a clean skill reads B.
What can Defi Overview access on my machine?
The audit observed no filesystem, network or shell use at all in its source.
How current is this page?
The grade is for one exact copy of the source (0448e87653f9), read on 2026-10-14. The repository is watched, and a new audit runs when it changes — this is the first audit.

Sources

Spotted something wrong? Report a correction · Methodology

Keep going in your AI

Opens with this page as the source. We add nothing else to the prompt.