Skip to content

Releases

v0.1.0

Choose a tag to compare

@github-actions github-actions released this 06 Sep 19:05

A single-binary bug bounty engine — scope-gated recon, template + active detection, deterministic verification, and LLM-assisted triage in one static Go binary with zero runtime dependencies. Detection-only: it proves vulnerabilities, it does not exploit them.

Live demo: https://talkdedsec.github.io/tlk-huntx/

First public release.

Safety core

  • Scope gate — machine-readable in/out-of-scope decision (wildcards + CIDR,
    out-of-scope precedence) on every outbound request.
  • Per-host rate limiter with global concurrency cap and 429/WAF back-off.
  • Blast-radius guard — total-request budget and a block on state-changing methods
    unless explicitly opted in.
  • Authorization gate — active pipelines refuse to run without a scope file and
    --authorized. The fetch client is the only component allowed to touch a target.

Discovery

  • recon — passive subdomain discovery (crt.sh, Wayback, OTX, HackerTarget) plus
    gated HTTP probing and an attack-surface graph.
  • crawl — mine endpoints and parameters from HTML, same-host JavaScript,
    robots.txt Allow/Disallow, and sitemap.xml.
  • fuzz — content discovery over a built-in common-paths wordlist (soft-404 aware).
  • api — GraphQL introspection and OpenAPI/Swagger discovery.
  • Session identities for authenticated crawling and authz testing.

Detection

  • scan — nuclei-compatible template engine with built-in templates, tech-aware
    selection, and secret scanning.
  • Active, detection-only checks: reflect/xss, sqli, ssti, lfi,
    redirect, ssrf, cmdi, hostheader.
  • bola/idor — object-level authorization testing across named identities.
  • greybox — correlate dangerous source sinks with live endpoints.

Verification

  • Deterministic re-confirmation out-of-band; the engine proves a finding before a
    model ever sees it.
  • Built-in OOB collaborator for SSRF/command-injection callbacks.
  • BOLA/IDOR differential checks.

Intelligence

  • plan — rank findings by expected value and compose known attack chains.
  • feedback — record triage decisions and self-calibrate confidence.
  • monitor — diff a run against the last; --interval to repeat, --notify webhook.
  • mcp — Model Context Protocol server (stdio) exposing scope/recon/scan/plan as
    typed tools, plus the /huntx skill for LLM-driven triage and reporting.

Interface

  • full — one-shot pipeline: recon + crawl + scan + active checks + ranked plan.
  • dashboard — interactive local console with a finding detail drawer, severity
    filters, search, and a Findings/Hosts/Surface (attack-surface graph) view.
  • report — HackerOne/Bugcrowd-ready Markdown and SARIF 2.1.0 output;
    --min-severity to filter.
  • coordinate/worker — distributed scanning over HTTP.
  • Styled ANSI startup banner (NO_COLOR aware).

Install

go install github.com/talkdedsec/tlk-huntx/cmd/huntx@v0.1.0

Or grab a pre-built binary for your platform below (each archive ships an SBOM; verify with checksums.txt).