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-business-logic-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-business-logic-flaws)<a href="https://agentmods.dev/skills/unboundcompute/security-agent-skills/hunting-business-logic-flaws"><img src="https://agentmods.dev/badge/skills/unboundcompute/security-agent-skills/hunting-business-logic-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-business-logic-flaws"><img src="https://agentmods.dev/badge/skills/unboundcompute/security-agent-skills/hunting-business-logic-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 | $0.00109 | $0.01575 |
| Opus 5 | $0.00055 | $0.00788 |
| Sonnet 5 | $0.00022 | $0.00315 |
| Haiku 4.5 | $0.00011 | $0.00158 |
Grade A, and why
hunting-business-logic-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 7d 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 — 133 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Hunting business-logic flaws: the bugs where every request is valid
A business-logic flaw is a gap between what the code enforces and what the business intends. No request is malformed, no payload is injected, no signature is invalid; the attacker just does allowed things in a disallowed order, quantity, or combination. Scanners cannot find these, because there is no dangerous function call to flag. The danger is the sequence and the arithmetic, so finding them means modeling the intended rule first, then searching for a permitted path that breaks it.
When to use
- You are reviewing checkout, payment, transfers, refunds, or redemption.
- A multi-step flow with a required order (signup, approval, provisioning).
- Anything with a quota, a limit, a balance, a tier, or a one-time action.
Scope check
Test flows you own or are authorized to test. Exercise sequences against your own accounts and data; do not move real money or affect real users. If you can't name the authorization, stop.
The loop
-
Recover the intended rules. For the flow under review, write down the invariants the business assumes: "a coupon applies once," "you cannot ship what you did not pay for," "a transfer cannot exceed the balance," "step 3 requires step 2." These are usually unwritten. State each explicitly, because each is a hypothesis you will try to break.
-
Map the state machine and its real transitions. Diagram the states and the requests that move between them. Then find transitions the diagram forbids but the code permits: can you post the final step directly, replay an approval, revisit a completed state, or reach a state without its precondition? Missing server-side enforcement of order is step-skipping.
-
Attack the arithmetic and the limits. For every quantity, price, count, or balance the user influences, test the edges: negative, zero, very large, fractional and rounding boundaries, and the value that pushes a cumulative total past a cap. A quantity that can go negative credits the attacker; a limit checked before but not after a concurrent change is a limit-overrun.
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.
- 7d ago First seen · 133 lines · 109 tokens per session scan A 6ff81db2ab91
hunting-business-logic-flaws is a skill published in the GitHub repository UnboundCompute/security-agent-skills (5 stars, last pushed 3d ago), licensed MIT. It adds 109 tokens to every session and 1,575 once invoked, about $0.0005 per session on Opus 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
ffuf-claude-skill
Skill "ffuf-claude-skill" from Agent-Threat-Rule/agent-threat-rules, covering ffuf claude skill, when to use this skill and instructions.
codeinspectus-fix-one
Investigate and remediate exactly one user-selected CodeInspectus finding with evidence-gated reproduction, a separately approved minimal patch, focused regression testing, and an exact-prior-scan rescan. Use when a user asks an agent to examine, reproduce, fix, or verify one CodeInspectus finding without batching…
systematic-debugging
Use when encountering any bug, test failure, or unexpected behavior, before proposing fixes.
gauntlex:doctor
Full GAUNTLEX environment health check — model reachable, ChromaDB writable, AVF fixtures pass, no unexpected outbound network (air-gap verification). Use before first run or when debugging failures.
diagnose-bundle
Maintainer/dev skill that triages an anonymised diagnostic bundle (appsec-diag-.tgz, produced by scripts/diagnosticbundle.py) a user sent after a pipeline failure. Runs the deterministic inspect, then cross-references the plugin source (scripts/, compact runtimes, agents/, AGENTS.md) and known-bug history to produce a…
diagnose-run
Find the plugin defects behind the issues a create-threat-model run recorded. Reads .run-issues.json, re-reads each symptom next to the plugin's own code, and reports a file:line root cause per issue, or classifies it as an environment or expected condition. Read-only against the plugin; writes only .run-bugs.json.…