Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx skills add UnboundCompute/security-agent-skills --skill hunting-defi-economic-and-oracle-flawsgit clone --depth 1 https://github.com/UnboundCompute/security-agent-skillsWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/skills/unboundcompute/security-agent-skills/hunting-defi-economic-and-oracle-flaws)<a href="https://agentmods.dev/skills/unboundcompute/security-agent-skills/hunting-defi-economic-and-oracle-flaws"><img src="https://agentmods.dev/badge/skills/unboundcompute/security-agent-skills/hunting-defi-economic-and-oracle-flaws/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/skills/unboundcompute/security-agent-skills/hunting-defi-economic-and-oracle-flaws"><img src="https://agentmods.dev/badge/skills/unboundcompute/security-agent-skills/hunting-defi-economic-and-oracle-flaws.svg" alt="Reviewed on agentmods" width="80" height="20"></a>What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00191 | $0.02013 |
| Opus 5.5 | $0.00076 | $0.00805 |
| Sonnet 5.5 | $0.00038 | $0.00403 |
| Haiku 4.5 | $0.00019 | $0.00201 |
Grade A, and why
hunting-defi-economic-and-oracle-flaws scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 1mo ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 130 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Hunting DeFi economic and oracle flaws: profiting by moving a price nobody bounded
Some of the most valuable protocol bugs break no access check and re-enter nothing: they move a price or exploit accounting math to extract value through actions the protocol permits. A contract reads a spot price a flash loan can swing inside one transaction, trusts a pool ratio an attacker sets, or rounds a fee in a direction a repeated call drains. This hunt is about the economic layer, the values a protocol trusts and the invariants it assumes hold. It is distinct from access control (nothing here requires a missing gate) and from reentrancy (nothing here requires re-entry); if the fix is a manipulation-resistant oracle, a sanity bound, or corrected math, it is an economic flaw and belongs here. The discipline in a crowded field is proving the manipulation is profitable after costs, not just that a price can move.
When to use
- A protocol reads a price, a pool ratio, or a share value to make a swap, loan, or liquidation decision.
- Fee, rounding, or share math runs on values an attacker can influence within a transaction.
- You want to know whether an attacker can profit by moving a price or breaking an accounting invariant.
Scope check
Analyze only protocols you own or are authorized to assess, and simulate an economic exploit only on a fork or test network, a manipulation here extracts real value from real liquidity providers. Prove profitability on a fork. If you can't name the authorization, stop.
The loop
-
Find the value decisions and their inputs first. List the points where the protocol reads a price, a ratio, or a share value to move funds (a swap, a loan, a liquidation, a mint or redeem), and for each trace the input it trusts and whether an attacker can move that input within a transaction. The economic bug lives at a value decision that trusts a movable input; establish those before anything else.
-
Check the price source. Look for a price read from a spot source an attacker can move in the same transaction (a single pool's instantaneous ratio), a single-source oracle with no sanity bound or staleness check, or a feed that can be pushed outside a sane range without the protocol noticing.
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 1mo ago First seen · 130 lines · 191 tokens per session scan A 6fb615661a27
hunting-defi-economic-and-oracle-flaws is a skill published in the GitHub repository UnboundCompute/security-agent-skills (5 stars, last pushed 1mo ago), licensed MIT. It adds 191 tokens to every session and 2,013 once invoked, about $0.0008 per session on Opus 5.5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-09-05.
Other skills, from other repositories
insecure-openclaw
Track your crypto holdings across exchanges.
prodcheck-review
Review this codebase against the prodcheck pre-production checklists — security, performance, scale, integrations, post-launch readiness and people. Use when asked to check whether a project is ready to ship, to audit an area before launch, or to work through a specific checklist. Produces evidence with file:line…
integrate-arcjet-guard-genkit
Integrate Arcjet security into a Genkit JS agent using @arcjet/guard — wrap ai.defineTool, put guardMiddleware on generate({ use }) for unwrapped / MCP / filesystem tools, and read a caller-owned id from generate({ context }). Use when asked to add Arcjet to genkit, rate limit its tools, screen inbound messages, or…
integrate-arcjet-guard-tanstack-ai
Integrate Arcjet security into a TanStack AI chat() app using @arcjet/guard — put guardMiddleware first on chat({ middleware }) so onBeforeToolCall gates tools, and read a caller-owned id from chat({ context }). Use when asked to add Arcjet to TanStack AI, rate limit its tools, screen inbound messages, or block prompt…
integrate-arcjet-guard-claude-managed-agents
Integrate Arcjet security into Claude Managed Agents (hosted REST+SSE, beta managed-agents-2026-04-01) using @arcjet/guard — screen user.message / initialevents before sessions.events.send, and gate custom tools on agent.customtooluse. Use when asked to add Arcjet to Claude Managed Agents, rate limit custom tools, or…
linear-claude-skill
Manage Linear issues, projects, and teams.