LitterBoxBLOCK
A self-hosted sandbox for red teams to test payloads against modern detection before deployment. MCP integration lets an LLM agent drive analysis end to end.
Overview
From the repository's own README, as read at the audited commit. Badges and raw HTML are left out.
[]() []() []() []() []() [](https://deepwiki.com/BlackSnufkin/LitterBox) [](https://github.com/BlackSnufkin/LitterBox/stargazers)
A self-hosted payload-analysis sandbox for red teams. Upload a sample, run static / dynamic / EDR analysis against it, get a Detection Score and a triggering-indicators breakdown — decide whether the payload is field-ready before it leaves the lab.
LitterBox can also dispatch payloads to a separate EDR-instrumented Windows VM (Elastic Defend or Fibratus) and pull the correlated detection alerts back into the results page.
While designed primarily for red teams, LitterBox is equally useful for blue teams running the same tools in their malware-analysis workflows.
Documentation
Operator and developer documentation lives in the LitterBox Wiki.
cd0acb3a1830OBSERVED · 2026-09-24Exposed tools (20)
15 read · 3 write · 2 destructive. Blast radius: 2 tools can delete or overwrite — an agent that can be talked into calling a tool can be talked into calling this one.
| Tool | Risk | Description |
|---|---|---|
analyze_fuzzy_similarity | read | Score a payload |
cleanup_sandbox | destructive | Wipe analysis artifacts. DESTRUCTIVE — confirm with the user before calling. |
compare_with_blender | read | Compare a payload |
delete_payload | destructive | Delete one payload and its results. DESTRUCTIVE — confirm with the user before calling. |
download_report | read | Download the HTML report to disk and return the saved path. |
get_comprehensive_results | read | All available results in one parallel call (file_info + static + dynamic + holygrail). |
get_dynamic_results | read | Dynamic analysis output (memory scanners, behavioral telemetry, process output). |
get_edr_agents_status | read | Live probe of every EDR profile (Whiskers agent + Elastic stack reachability, |
get_edr_index | write | Index of every saved EDR run for a target (one entry per profile that has data). |
get_file_info | read | File metadata: type, size, hashes, entropy, PE structure, sensitive imports. |
get_holygrail_results | write | HolyGrail BYOVD output for a driver (LOLDrivers / block status / critical imports). |
get_report | read | Render the full HTML analysis report and return it inline as a string. |
get_risk_assessment | read | Computed detection assessment: numerical score, level (Low / Medium / High / Critical), triggering indicators. |
get_scanners_status | read | Inventory of configured local analyzers (static + dynamic + holygrail) and |
get_static_results | read | Static analysis output (YARA matches, CheckPlz findings, Stringnalyzer indicators). |
list_edr_profiles | read | List EDR profiles registered under Config/edr_profiles/. |
list_payloads | read | List every analyzed payload, driver, and process in the sandbox with detection summary. |
run_blender_scan | write | Snapshot the live host so Blender can compare payload runtime indicators against it. |
sandbox_status | read | Health, tool readiness, and fleet summary for the LitterBox server. |
validate_pid | read | Confirm a PID exists and is accessible before targeting it for dynamic analysis. |
Trust audit
BLOCKgrade D · trust 69/100 Do not install this without reading the findings. The audit found something that could harm you or your machine.
| Layer | What it checks | Result |
|---|---|---|
| L0 | Provenance & inventory | WARN |
| L1 | Static analysis of the code | FAIL |
| L2 | Instruction surface (what it tells the agent) | PASS |
| L3 | Class-specific surface | WARN |
| L4 | Behavioural (sandbox) | SKIPPED |
What the source does
- Filesystem
- declared (7 observation(s))
- Network
- declared (11 observation(s))
- Shell
- declared (2 observation(s))
- Dependencies
- pinned
- Secrets in source
- none-found
Findings (25)
session.verify = False
CheckPlz.exe
HolyGrail.exe
Hunt-Sleeping-Beacons.exe
Moneta64.exe
Patriot.exe
ctx.element.innerHTML = cleanState('No sleep-pattern indicators', `Process ${processName} (PID ${processPid}) shows no beacon-like sleep behaviour.`);agent_url: "http://192.168.1.100:8080"
elastic_url: "https://192.168.1.50:9200"
agent_url: "http://192.168.1.100:8080"
client = LitterBoxClient(base_url="http://127.0.0.1:1337", logger=logger)
'isdebuggerpresent', 'isprocessorfeaturepresent', 'loadlibrarya',
cleanup_sandbox, delete_payload
md5_hash = hashlib.md5()
'md5': hashlib.md5(self.indata).hexdigest(),
'sha1': hashlib.sha1(self.indata).hexdigest(),
md5_hash = hashlib.md5(file_content).hexdigest()
import { escapeHtml } from '../../utils/escape.js';python litterbox.py --debug # http://127.0.0.1:1337
<img class="brand-logo" src="data:image/png;base64,AAABAAYAEBAAAAAAIACiAgAAZgAAACAgAAAAACAA4AYAAAgDAAAwMAAAAAAgACMMAADoCQAAQEAAAAAAIAA1EgAACxYAAICAAAAAACAAbC0AAEAoAAAAAAAAAAAgAFxuAACsVQAAiVBORw0KGgoAA
['Has atob()', f.has_atob],
notes.append("atob() + Uint8Array decode chain present")Scanners/HollowsHunter/hollows_hunter.exe
Scanners/HolyGrail/Policies/lol_drivers.json
Scanners/PE-Sieve/pe-sieve.exe
Gates applied: no_behavioural_pass.
cd0acb3a1830full audit observations/trust-audit/mcp-server/blacksnufkin__litterbox.json · Report an issue / request a re-scanAudit history
Every audit this server has had. A grade with a past is a grade somebody is still checking.
| Date | Source | Verdict | Grade | Score | Change |
|---|---|---|---|---|---|
| 2026-09-24 | cd0acb3a1830 | BLOCK | D | 69 | first audit |
Questions
What is the LitterBox MCP server?
A self-hosted sandbox for red teams to test payloads against modern detection before deployment. MCP integration lets an LLM agent drive analysis end to end.
What tools does LitterBox expose?
20 in total: 15 read-only, 3 that write, and 2 that can delete or overwrite (cleanup_sandbox, delete_payload). Every one is listed on this page with its risk.
Is LitterBox safe to connect to an agent?
No — not without reading the findings first. The audit graded it D (69/100) and found 1 critical or high issue in the source. Each one is listed on this page with the file and line it is on. Separately from the audit: 2 of its tools can destroy data, so scope the token you give it to what you actually need.
What credentials does LitterBox need?
No credential environment variables were found in its source, so it appears to need none.
How does LitterBox run?
It speaks stdio and streamable-http, so it runs as a local process your client starts.
How current is this page?
The grade is for one exact copy of the source (cd0acb3a1830), read on 2026-09-24. The repository is watched and re-audited when it changes.