Skip to content
Security
Skill

/disclosure

Drive responsible disclosure of a proven finding to a CVE. Package the report, find the vendor contact, report privately, coordinate a timeline, request the CVE (vendor CNA / GitHub / MITRE), and publish an advisory. Closes the research loop. Triggers - "disclose", "request a

BOOST
From plugin
torch
33356 skills
Install
$ npx -y skills add Encod3d-Sec/TORCH --skill disclosure --agent claude-code

How it fires

How this skill gets triggered: by you, by Claude, or both.

  • Fires itselfAuto-invocation. Claude auto-loads it when your prompt matches the work.Auto-invocation is when the right skill fires by itself at the right moment, driven by a FLOW.md router and a hook, instead of you invoking it by name. It is the difference between a skill being installed and a skill actually getting used.Read the full definition →
  • You can call itInvoke it directly when you want it.
  • Slash command/disclosure

Context preview

The summary Claude sees to decide when to auto-load this skill.

Drive responsible disclosure of a proven finding to a CVE. Package the report, find the vendor contact, report privately, coordinate a timeline, request the CVE (vendor CNA / GitHub / MITRE), and publish an advisory. Closes the research loop. Triggers - "disclose", "request a

SKILL.md

disclosure.SKILL.md
name: disclosure
description: Drive responsible disclosure of a proven finding to a CVE. Package the report, find the vendor contact, report privately, coordinate a timeline, request the CVE (vendor CNA / GitHub / MITRE), and publish an advisory. Closes the research loop. Triggers - "disclose", "request a cve", "report this to the vendor".

Disclosure: Finding -> CVE

Turn a proven, novel vulnerability into a coordinated disclosure and a published CVE. Pairs with the `research` skill (a finding in `findings.md`) or an engagement FIND. Read [[responsible-disclosure]] first.

Gate (READ FIRST)

  • The finding must be **proven + reproducible** (minimal PoC, affected versions, impact) and **novelty-checked** (not an existing CVE). If not, go back to the `research` skill.
  • Confirm you are **authorised / in safe-harbor** for the target. No authorization -> do not proceed (CFAA / Computer Misuse Act). Research on public software you can lawfully analyze is fine; testing live third-party systems needs permission.

Procedure

1. **Package the report** from the finding: title, affected + fixed/tested versions, CWE class, CVSS vector + score, clear repro steps, minimal PoC, impact, suggested remediation, your contact. Redact any real data. 2. **Find the contact** (in order): `/.well-known/security.txt`, `SECURITY.md`, `security@<vendor>`, vendor PSIRT, a bug-bounty platform if they run one, GitHub private vulnerability reporting for OSS, else CERT/CC. 3. **Report privately** over that channel; PGP-encrypt if a key is published. Professional, specific, non-extortive. Offer to validate the fix. 4. **Coordinate a timeline** - propose ~90 days; track it; escalate to CERT/CC if the vendor goes dark past the deadline. 5. **Request the CVE:** vendor CNA assigns it; for OSS open a GitHub Security Advisory (GitHub issues the CVE); otherwise MITRE CVE request form (CNA of last resort). Supply product, versions, CWE, impact, reference. 6. **Publish the advisory** after fix/deadline: CVE ID, affected/fixed versions, CWE, CVSS, description, PoC, impact, remediation, timeline, credit. Mirror to the researcher's blog/GitHub for the portfolio.

Output

  • A ready-to-send **disclosure report** (draft) and a **public advisory** (draft) saved under the project: `raw/research/<project>/advisory.md` (and the contact + timeline tracked in `findings.md`).
  • Update the finding status: candidate -> reported -> `CVE-<id>` -> published.

Wiki feedback

Reusable disclosure lesson (vendor process quirk, CNA tip) -> update [[responsible-disclosure]]. The vuln technique itself -> the matching `wiki/techniques/` page via `research-ingest`.

Report: report + advisory drafts, contact channel, and the disclosure timeline.

Read more
Ships withtorch

An AI-powered penetration testing and bug-bounty knowledge base and automation harness for Claude Code.

Get the whole plugin

Other skills on torch.

hunt-burp
Skill

hunt-burp

Drive Burp Suite over its MCP server as an AI triage + attack layer - review proxy history for signals, replay via Repeater/send,…

hunt-ad
Skill

hunt-ad

Active Directory attack hunting - enumeration to domain dominance. Spray-safe (lockout gate), AS-REP/Kerberoast, ACL + ADCS…

hunt-api
Skill

hunt-api

API attack hunting (REST / GraphQL / gRPC) - BOLA/IDOR, BFLA, mass assignment, excessive data exposure, auth/JWT, introspection +…

hunt-auth
Skill

hunt-auth

Auth bypass and ATO hunting - legacy protocol matrix (XMLRPC, SharePoint /_vti_bin/, EWS, Citrix, etc.), JWT manipulation,…