Last checked with GitLab: . Status comes from GitLab using a 2-minute cache. Merge events clear affected cached statuses. An already-open page does not update itself. Refresh page.

An MR completes this bounty when it is merged into main with Closes #293 and GitLab closes the issue. A mention or an updated timestamp is not completion. Awarded points appear in Activity.

Description

The Purple Team guide defines detected, visible-not-alerting, invisible and untested outcomes. Make that teaching model executable using original synthetic log/alert evidence, without running attacks.

Who might enjoy this

  • Community interests: Purple team, Research, Tech
  • Skills: python, qa, security
  • Experience: intermediate; work style: automation.
  • Readiness: ready; bounty size: M (existing points tier; standard merge award stacks separately).

Current behavior and starting points

Source-reviewed against main cf05b29c6256dba70d0ebc09b22cc9867cdb786a on2026-10-18. Check current main and related issues/MRs before implementing so existing work is reused.

Relevant Files

Deliverables

  • A bounded JSONL synthetic evidence corpus and expected-results manifest for three fictional exercise markers.
  • An offline checker/report that separates signal presence, alert presence, coverage state and missing evidence; include original exercise walkthrough.

Dependencies and decisions

No prerequisite issue is required. Re-read current main and linked MRs before starting.

Acceptance criteria

  • [ ] A known positive produces detected, a log-only case produces visible-not-alerting, and an explicit no-signal case remains distinct from untested.
  • [ ] Missing required evidence or malformed event time yields unavailable/invalid, never a passing detection result.
  • [ ] Mutating/removing an expected synthetic alert makes the corresponding regression fail; no command executes an attack, contacts hosts or inspects production logs.
  • [ ] Record named baseline failures and the exact commands used. A behavior fix includes a real route/database/protocol check observed failing before the change. Knowledge work includes source traceability and reproducible reviewer evidence.
  • [ ] Run the affected app's existing checks. Explain any baseline failures rather than reporting a matching failure count as proof.

Try it locally

Start from the existing source files above. The following is the target demo your contribution should enable; a newly proposed script or fixture is part of the deliverable:

Replay the original synthetic event/alert files and remove an alert from a copy to see the coverage result change.

Use public or synthetic examples. Starter work should run without production access; explicitly name any software/device prerequisite before claiming it is ready.

Out of scope

  • No vulnerability hunting or publication, real adversary emulation, exploit execution, live SIEM access or production log export.
  • Do not duplicate #101-111 broad bug-bounty targets or #257 application access-matrix tests.
  • Unrelated refactors, new services/providers, bulk dependency changes, production access changes, unrequested notifications, and copying private member material. Keep credentials and unresolved security findings outside the issue and repository.

For a coding agent and reviewer

Read AGENTS.md, applicable .cursor/rules/, app instructions and docs/contributing/agentic-bounty-review.md. Reuse the existing shared helpers, identity/access paths and app styling. Implement one bounded story and provide evidence the accepting app maintainer can reproduce. State updated / pre-release verified / released / live verified accurately.

A parent index carries no duplicate award for these child deliverables. Open one scoped MR using this issue's actual number after the closing keyword. Link your GitLab identity on Scoreboard; the existing merge webhook handles eligible points.

Catalog index: #258.

<!-- community-interest-bounty:2026-10-18:purple-coverage-replay -->

Relevant files

Hints

No hints yet.

Activity

  • +100 Hugo Zarate earned M bounty Oct 25

Comments

No comments yet. Be the first to weigh in.

↑ Top