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 forefy/.context --skill hackerone-reportgit clone --depth 1 https://github.com/forefy/.contextWrote 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/forefy/.context/hackerone-report)<a href="https://agentmods.dev/skills/forefy/.context/hackerone-report"><img src="https://agentmods.dev/badge/skills/forefy/.context/hackerone-report.svg" alt="Measured on agentmods" height="20"></a>- NVIDIA SkillSpector pass
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 | $0.00039 | $0.01923 |
| Opus 5 | $0.00019 | $0.00962 |
| Sonnet 5 | $0.00008 | $0.00385 |
| Haiku 4.5 | $0.00004 | $0.00192 |
Grade A, and why
hackerone-report 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 4d 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 — 83 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Filing a HackerOne report
Drive a HackerOne vulnerability submission end to end: verify it's worth filing, write it up concisely, build a reproducible PoC with a video-ready README, fill the submission form in the browser, and hand the irreversible steps back to the user.
Non-negotiable safety rules
These are hard stops. They protect the user's account and their Signal.
- Never enter credentials or 2FA codes. When the form gates on login, take a screenshot, tell the user to sign in themselves, and wait. Suggest they tick "Remember me for 2 weeks" (HackerOne sessions drop mid-flow otherwise).
- Never click the final "Submit Report". It is an irreversible publish and it spends one of the user's reports. Fill everything, then hand off: the user reviews and submits. State this explicitly up front.
- Never attach files the user didn't ask you to, and don't fabricate PoC output. The in-app browser can't drive the native file picker anyway - the user attaches the zip and the video.
- Filing is per-user consent. Draft freely; the act of submitting is the user's.
Phase 0 - Dedup pre-flight (do this FIRST, every time)
The most expensive mistake is re-filing a finding the program already closed. It wastes a report and dents Signal. Before drafting anything:
- Search the local findings tree for prior work on the same root cause / sinks / code paths. Look especially in any
duplicates/folder and for filenames encoding a resolution:Informative-*,Duplicate-*,NA-*,Resolved-*. A file namedInformative-1-<x>.mdmeans the program already closed<x>as Informative. - Compare code locations, not just titles. Grep both the candidate writeup and prior reports for the exact files/functions/sinks cited. If they overlap, it's the same finding regardless of new framing. (A "new vector" that strengthens the entry but reaches the same sink the program already judged is still a duplicate.)
- Offer to check the live program state: open HackerOne → My Reports, search a keyword from the finding, read the actual close reason/comment.
- If it's a dup: say so plainly and recommend not filing. Options: don't file; comment/appeal on the existing report instead of opening a new one; hold reports for something novel. Only proceed on the user's explicit go-ahead.
- Also surface program constraints visible on the submit page: "Trial Reports Remaining: N" and any Signal Requirement - a low Signal can block future submissions, so accuracy matters more than volume.
What ships with it
2 files 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.
- 4d ago Changed · +1 lines cb202f77fda6
- 8d ago First seen · 82 lines · 39 tokens per session scan A 885547b7739a
hackerone-report is a skill published in the GitHub repository forefy/.context (144 stars, last pushed yesterday), licensed MIT. It adds 39 tokens to every session and 1,923 once invoked, about $0.0002 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 2026-08-30.
Other skills, from other repositories
web2-vuln-classes
Complete reference for 26 web2 bug classes with root causes, detection patterns, bypass tables, exploit techniques, and real paid examples. Covers IDOR, auth bypass, XSS, SSRF (11 IP bypass techniques), SQLi, business logic, race conditions, OAuth/OIDC, file upload (10 bypass techniques), GraphQL, LLM/AI (ASI01-ASI10…
asc-app-create-ui
Create a new App Store Connect app record via browser automation. Use when there is no public API for app creation and you need an agent to drive the New App form.
agent-browser
Browser automation for AI agents via inference.sh. Navigate web pages, interact with elements using @e refs, take screenshots, record video. Capabilities: web scraping, form filling, clicking, typing, drag-drop, file upload, JavaScript execution. Use for: web automation, data extraction, testing, agent browsing…
agent-browser
Browser automation CLI for AI agents. Use when the user needs to interact with websites, including navigating pages, filling forms, clicking buttons, taking screenshots, extracting data, testing web apps, or automating any browser task. Triggers include requests to "open a website", "fill out a form", "click a…
agent-canvas
Interactive element picker for web pages. Opens a browser with click-to-select UI overlay. Use when you need to let users visually select DOM elements, identify element selectors, or get detailed element information interactively. Triggers on "select an element", "pick element", "let me choose", "which element", or…
agent-eyes
Visual context analyzer for AI agents. Provides screenshots, accessibility scans, DOM snapshots, and element descriptions for web pages. Use when you need to see what a web page looks like, analyze accessibility issues, inspect DOM structure, or get detailed element information. Triggers on requests like "take a…