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 SecurityTalent/bugskill-ai --skill request-smugglinggit clone --depth 1 https://github.com/SecurityTalent/bugskill-aiWrote 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/securitytalent/bugskill-ai/request-smuggling)<a href="https://agentmods.dev/skills/securitytalent/bugskill-ai/request-smuggling"><img src="https://agentmods.dev/badge/skills/securitytalent/bugskill-ai/request-smuggling/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/securitytalent/bugskill-ai/request-smuggling"><img src="https://agentmods.dev/badge/skills/securitytalent/bugskill-ai/request-smuggling.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 | $0.00062 | $0.03178 |
| Opus 5.5 | $0.00025 | $0.01271 |
| Sonnet 5.5 | $0.00012 | $0.00636 |
| Haiku 4.5 | $0.00006 | $0.00318 |
Grade B, and why
request-smuggling scanned grade B 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 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.
Sends data to an external URLmediumData exfiltration
A POST to an outside endpoint may be telemetry or may be exfiltration; either way the mod talks to somewhere, and you should know where.
fetch("https://target.example/vulnerable", { method: "POST", This is a copy
100% identical to request-smuggling — 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 — 315 lines — stays where its author put it; the contents beside it link to each section on GitHub.
SKILL: HTTP Request Smuggling — Expert Attack Playbook
AI LOAD INSTRUCTION: Expert HTTP desync techniques. Covers CL.TE, TE.CL, TE.TE obfuscation variants, HTTP/2 downgrade and pseudo-header confusion, client-side desync (browser
fetchpipelines), and tool-assisted fuzzing. Assumes familiarity with raw HTTP/1.1 framing and reverse-proxy topologies. This is not “header injection” — it is message boundary disagreement between hops.
Routing note: load this skill when you suspect CDN/reverse-proxy and origin disagree on request-end boundaries, or when abnormal concatenation appears during H2-to-H1 downgrade.
0. RELATED ROUTING
- ghost-bits-cast-attack when the HTTP client library is Apache HttpClient <= 4.5.9 (HTTPCLIENT-1974/1978) — injecting
瘍瘊(U+760D U+760A, low bytes\r\n) into a header value causes the underlying char-to-byte writer to emit a literal CRLF, splitting the request at the origin without relying on CL/TE disagreement
1. QUICK START
CL.TE first probe (front-end trusts CL, back-end trusts chunked)
Assumption: front end prioritizes Content-Length, back end prioritizes Transfer-Encoding: chunked. Use a very short CL so the front end accepts a fake end, while the back end continues chunk parsing and leaves remaining bytes for the next request.
POST / HTTP/1.1
Host: target.example
Content-Type: application/x-www-form-urlencoded
Content-Length: 13
Transfer-Encoding: chunked
0
SMUGGLED
- Front end reads only 13 bytes based on
Content-Length: 13(that is,0\r\n\r\nSMUGGLED, 13 bytes total) and considers the request complete. - Back end parses as chunked: after the
0end chunk, it treatsSMUGGLEDand onward as the start byte stream of the next request.
TE.CL first probe (front-end trusts chunked, back-end trusts CL)
Assumption: front end parses chunked and back end only reads Content-Length. Set CL equal to the number of bytes in the chunk-length line (commonly 4: two hex characters + \r\n), so the back end consumes only the length line and leaves the rest buffered for follow-up request splicing.
What ships with it
1 file beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
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 · 315 lines · 62 tokens per session scan B 6469b8e00d95
request-smuggling is a skill published in the GitHub repository SecurityTalent/bugskill-ai (13 stars, last pushed 21d ago), licensed MIT. It adds 62 tokens to every session and 3,178 once invoked, about $0.0002 per session on Opus 5.5. A static security scan graded it B with 1 finding (sends data to an external url). It is 100% identical to request-smuggling, differing in 0 lines, and is treated as a copy.
Other skills, from other repositories
user-research-cookiy
End-to-end user research assistant — qualitative and quantitative. Use this skill whenever the user mentions user research, user interviews, discussion guides, interview guides, research plans, qualitative research, quantitative research, user surveys, survey design, usability studies, participant recruitment…
email-marketing-bible
Email marketing operating manual for AI agents, with hard send-safety gates. Use when someone plans, writes, designs, audits or sends an email campaign, newsletter, welcome series, abandoned-cart, post-purchase or win-back flow; builds a list or segment; drives Klaviyo, Mailchimp, Brevo, Kit or any ESP through MCP, an…
globalpercent
GlobalPercent — build a global-macro-probability panel for an investment research system. Merges public probability data from prediction markets (Polymarket + Kalshi), classifies every market into macro modules (monetary policy / macro economy / AI / etc.), and shows the whole market's expected-probability state at a…
ui-ux-design-pro
Senior-level UI/UX design expert for building data-driven, premium production interfaces. Use when you need to: 1. Design complex applications (dashboards, SaaS, AI tools) from scratch 2. Generate comprehensive design systems (tokens, palettes, typography) 3. Audit existing UI for quality, accessibility, and "craft"…
playwright-best-practices
Use when writing Playwright tests, fixing flaky tests, debugging failures, implementing Page Object Model, configuring CI/CD, optimizing performance, mocking APIs, handling authentication or OAuth, testing accessibility (axe-core), file uploads/downloads, date/time mocking, WebSockets, geolocation, permissions…
open-map-stack
Use textual agent instructions for GIS and geospatial work: source discovery and provenance, vector/raster/point-cloud pipelines, CRS and metric analysis, spatial SQL, routing and isochrones, QGIS projects, tile generation, and web maps. Use advanced tools and formats such as OSM, Overture, STAC, Sentinel/Landsat…