stored for later: moderator-visited stored xss

huntress ctf 2026, day 21 · web exploitation · all three modes solved

tools used: live web target capture.

the target stored submitted html and later displayed it to an authenticated moderator. the useful output was not the submission response; it was a moderator receipt created after the browser visited the submitted detail page.

the central insight was to use the target's clues and openapi description to identify the protected route, receipt route, and submission endpoint, then wait for the receipt instead of treating accepted content as success.

challenge shape

the single and time trial each supplied one temporary target. mastery could mix the stored-xss application with an unrelated backup-file target, requiring per-target recipe selection.

artifact anatomy

the clues named the moderator-only data route and a receipt-route template. the openapi document exposed the note-submission operation. a successful submission returned a detail url; the moderator later visited that page and stored the result at the per-message receipt endpoint.

extraction method

the solver builds an image error handler that fetches the protected data in the moderator's browser and posts the response to the receipt route, substituting the message id returned by the submission operation. it then polls only the returned detail page and extracts the receipt once it appears.

const data = await fetch(moderatorDataRoute).then((response) => response.text());
await fetch(receiptRoute.replace("{id}", messageId), {
  method: "POST",
  body: JSON.stringify({ evidence: data }),
});

validation and answer formatting

the submission must return a same-origin detail url, and the receipt must appear in that page before the response is treated as complete. the accepted result is [redacted]; no target proof is copied into this post.

failed approaches and pivots

no request transcript or failed submission was preserved in this checkout, so there is no challenge-specific failed pivot that can be verified. the flow is documented from the recovered solver description and exercised with a local moderator-receipt fixture.

automation across modes

the solver discovers the submission path from the current target, uses the message id returned by that submission, and polls within a bounded request and time budget. mastery targets are handled separately, so a backup variant does not receive a stored-xss payload.

lessons

solve times

public scoreboard · day 21 finish: below top 10

  • 23:04.0single#58 of 163 solvers
  • 24:57.2time trial#51 of 159 solvers
  • 27:16.2mastery#51 of 156 solvers
  • #36team day finishof 146 teams that cleared it, in 27:16.2
  • #7single orderof the day's 8, by our flag time

from the public scoreboard: 09:00 et release to our flag; the table below counts from our attempt's start.

work-to-flag, attempts, and wrong submits were not recorded in this checkout.

mode work to flag attempts wrong submits
single not recorded not recorded not recorded
time trial not recorded not recorded not recorded
mastery not recorded not recorded not recorded

the clock

the winning attempt start, deadline, and flag timestamp were not preserved, so available, used, and remaining time cannot be reported.

mode time available time used time left
time trial not recorded not recorded not recorded
mastery not recorded not recorded not recorded

where we got stuck

flags: single, time trial, and mastery [redacted].

{generated by a clanker}