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 JustineDevs/premortem --skill open-redirectgit clone --depth 1 https://github.com/JustineDevs/premortemWrote 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/justinedevs/premortem/open-redirect)<a href="https://agentmods.dev/skills/justinedevs/premortem/open-redirect"><img src="https://agentmods.dev/badge/skills/justinedevs/premortem/open-redirect/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/justinedevs/premortem/open-redirect"><img src="https://agentmods.dev/badge/skills/justinedevs/premortem/open-redirect.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 | $0.00033 | $0.02930 |
| Opus 5 | $0.00016 | $0.01465 |
| Sonnet 5 | $0.00007 | $0.00586 |
| Haiku 4.5 | $0.00003 | $0.00293 |
Grade C, and why
open-redirect scanned grade C with 2 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 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.
Cloud metadata endpointhighServer-side request forgery
One request to 169.254.169.254 can return temporary IAM credentials.
# attacker.com returns 302 → http://169.254.169.254/ Makes network callslowCapability
Not a fault in itself. Listed so you know the mod talks to something, and to what.
Not all HTTP clients follow cross-protocol redirects, but `curl` (default) and some libraries do. This is a copy
100% identical to open-redirect — 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 — 381 lines — stays where its author put it; the contents beside it link to each section on GitHub.
SKILL: Open Redirect — Expert Attack Playbook
AI LOAD INSTRUCTION: Open redirect techniques. Covers parameter-based redirects, JavaScript sinks, filter bypass, and chaining with phishing, CSRF Referer bypass, OAuth token theft, and SSRF. Often underrated but critical for phishing and as a building block in multi-step exploit chains.
1. CORE CONCEPT
Open redirect occurs when an application redirects users to a URL derived from user input without validation. The trusted domain acts as a "launchpad" for phishing or token theft.
https://trusted.com/redirect?url=https://evil.com
→ User sees trusted.com in the link → clicks → lands on evil.com
2. FINDING REDIRECT PARAMETERS
Common Parameter Names
?url= ?redirect= ?next= ?dest=
?destination= ?redir= ?return= ?returnUrl=
?go= ?forward= ?target= ?out=
?continue= ?link= ?view= ?to=
?ref= ?callback= ?path= ?rurl=
Server-Side Sinks
HTTP 301/302 Location header
PHP: header("Location: $input")
Python: redirect(input)
Java: response.sendRedirect(input)
Node: res.redirect(input)
Client-Side (JavaScript) Sinks
window.location = input
window.location.href = input
window.location.replace(input)
window.open(input)
document.location = input
3. FILTER BYPASS TECHNIQUES
| Validation | Bypass |
|---|---|
Checks if URL starts with / |
//evil.com (protocol-relative) |
Checks domain contains trusted.com |
evil.com?trusted.com or trusted.com.evil.com |
Blocks http:// |
//evil.com, https://evil.com, \/\/evil.com |
Checks URL starts with https://trusted.com |
https://[email protected] (userinfo) |
Regex ^/[^/] (relative only) |
/\evil.com (backslash treated as path in some browsers) |
Django endswith('target.com') |
http://evil.com/www.target.com — URL path ends with target domain |
| Whitelist by domain suffix | Subdomain takeover on *.trusted.com |
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 · 381 lines · 33 tokens per session scan C ff08ad872fa5
open-redirect is a skill published in the GitHub repository JustineDevs/premortem (2 stars, last pushed 1mo ago), licensed Apache-2.0. It adds 33 tokens to every session and 2,930 once invoked, about $0.0002 per session on Opus 5. A static security scan graded it C with 2 findings (cloud metadata endpoint, makes network calls). It is 100% identical to open-redirect, differing in 0 lines, and is treated as a copy.
Other skills, from other repositories
opencli-autofix
Automatically fix broken OpenCLI adapters when commands fail. Load this skill when an opencli command fails — it guides you through collecting a trace artifact, patching the adapter, retrying, and filing an upstream GitHub issue after a verified fix. Works with any AI agent.
webapp-testing
Start/reuse a local app, wait for readiness, inspect rendered state/console/network, act from observed selectors, and verify with evidence.
agent-browser
Browser automation via Chrome/Chromium CDP — open, snapshot, click, screenshot. For testing web apps, mobile layouts, and automated interactions without Playwright/Puppeteer.
harbor-daytona
Use Harbor's Daytona sandbox platform for computer use — creating sandboxes, taking screenshots, sending mouse/keyboard input, and building agent loops. Use when the user wants to interact with a GUI, automate a desktop, do computer use, control a browser visually, or run Claude computer use against a Daytona sandbox.
web-scraper
Scrape, crawl, and extract data from websites. Use when users ask to scrape web pages, extract content, crawl websites, or collect data from the internet.
BrowserBash Browser Automation
BrowserBash is a vendor-independent, natural-language browser automation CLI. Drive a real browser from plain-English objectives or committable Markdown tests, run on local Chrome, CDP/Playwright MCP, Browserbase, LambdaTest, or BrowserStack, and stream NDJSON results with CI exit codes — using free local Ollama…