Getting it into your agent
It runs from inside its repository, so the clone comes first — what it calls does not travel with the file alone.
git clone --depth 1 https://github.com/akashrpatil/awesome-offensive-security-skillsnpx agentmods add skills/akashrpatil/awesome-offensive-security-skills/ai-report-writing-guardrailsWrote 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/akashrpatil/awesome-offensive-security-skills/ai-report-writing-guardrails)<a href="https://agentmods.dev/skills/akashrpatil/awesome-offensive-security-skills/ai-report-writing-guardrails"><img src="https://agentmods.dev/badge/skills/akashrpatil/awesome-offensive-security-skills/ai-report-writing-guardrails/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/akashrpatil/awesome-offensive-security-skills/ai-report-writing-guardrails"><img src="https://agentmods.dev/badge/skills/akashrpatil/awesome-offensive-security-skills/ai-report-writing-guardrails.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 | $0.00051 | $0.02846 |
| Opus 5 | $0.00026 | $0.01423 |
| Sonnet 5 | $0.00010 | $0.00569 |
| Haiku 4.5 | $0.00005 | $0.00285 |
Grade A, and why
ai-report-writing-guardrails 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 9d 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.
5. PoC: Working curl/script — ZERO placeholders This is a copy
100% identical to ai-report-writing-guardrails — 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 — 301 lines — stays where its author put it; the contents beside it link to each section on GitHub.
AI Report Writing Guardrails
When to Use
- When using Claude or any LLM to draft bug bounty vulnerability reports.
- When reviewing AI-generated reports before submission to HackerOne/Bugcrowd.
- When AI reports are getting marked as N/A, Informational, or Duplicate due to quality issues.
- When teaching Claude your personal writing style to produce better reports.
Prerequisites
- 3-5 of your best past bug bounty reports (ones that got triaged quickly and paid well)
- Claude Code CLI configured for the target program
- Understanding of CVSS 4.0 scoring
Core Problem: AI Reports Get You Banned
"AI blends 2-3 separate bugs into one report. Program managers HATE that. And the threat modeling is weak — it calls a paywall bypass 'complete security degradation'." — Critical Thinking Podcast, Ep. 166
The 3 Deadly AI Report Sins
| Sin | What Happens | Example |
|---|---|---|
| Bug Blending | AI mixes 2-3 separate vulnerabilities into 1 report | "Found XSS, CSRF, and IDOR in the dashboard" → should be 3 separate reports |
| Inflated Threat Model | AI exaggerates impact beyond technical reality | Paywall bypass → "complete security degradation of all financial systems" |
| AI Voice Detection | Triage team recognizes LLM language and downgrades | "It's worth noting", "This vulnerability poses a significant risk", "Certainly" |
Result: Report gets N/A'd, your reputation score drops, you may get warned or banned.
Workflow
Phase 1: The Bug Blending Check
Before submitting any AI report, apply the One Bug = One Report rule:
## Bug Blending Checklist
For each AI-generated report, ask:
1. How many DISTINCT attack flows are described?
- If > 1 → SPLIT into separate reports
2. Does each CWE map to the same root cause?
- Different CWEs = different reports (CWE-79 XSS ≠ CWE-352 CSRF)
3. Can each finding be independently reproduced?
- If yes → separate reports
- If chained (A enables B) → one report with clear chain
4. Remove sentences containing "Additionally", "Furthermore", "Moreover"
that introduce DIFFERENT vulnerability classes
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.
- 9d ago First seen · 301 lines · 51 tokens per session scan A bd06eb76506a
ai-report-writing-guardrails is a skill published in the GitHub repository akashrpatil/awesome-offensive-security-skills (5 stars, last pushed 4mo ago), licensed Apache-2.0. It adds 51 tokens to every session and 2,846 once invoked, about $0.0003 per session on Opus 5. A static security scan graded it A with 1 finding (makes network calls). It is 100% identical to ai-report-writing-guardrails, differing in 0 lines, and is treated as a copy.
Other skills, from other repositories
ai-report-writing-guardrails
Prevent common AI report pitfalls — bug blending, inflated threat models, and generic language. Train Claude with your best past reports for concise, technical submissions. Based on Critical Thinking Bug Bounty Podcast Episode 166.
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…
bug-bounty
Complete bug bounty workflow — recon (subdomain enumeration, asset discovery, fingerprinting, HackerOne scope, source code audit), pre-hunt learning (disclosed reports, tech stack research, mind maps, threat modeling), vulnerability hunting (IDOR, SSRF, XSS, auth bypass, CSRF, race conditions, SQLi, XXE, file upload…
credential-attack
Password spray methodology for bug bounty — when to do it vs web-vuln hunting, the wordlist-gen + breach-check + osint-employees + spray pipeline, mode selection (http-form / oauth / o365 / okta), rate-limit + lockout tactics, BBP legal guardrails, success detection, and the spray → authenticated /hunt chain pattern.…
graphql-audit
GraphQL security hunting — introspection abuse, field suggestion enumeration (clairvoyance), batching DoS, IDOR via aliasing, auth bypass, injection via arguments, subscription abuse, depth/complexity bombs, and WAF bypass. Covers graphw00f fingerprinting, gqlmap, graphql-cop, and inql. Use when a target exposes a…
mobile-pentest
Mobile app pentest for bug bounty (Android APK + iOS IPA) — runtime-first workflow: install app, proxy through Burp/mitmproxy, drive the UI, capture packets, then test the API exactly like a web target; escalate to decompile (apktool/jadx) and Frida/objection only when traffic is SSL-pinned, encrypted, or absent.…