Malicious PDF — malware analysis report

Static analysis result for SHA-256 6440b188949e2eaf…

MALICIOUS

PDF

45.2 KB
MD5: 8db9ac6d2254db6f79c0304fa123eaaf SHA-1: bfa59529d561d3fe9d67025af0bd65cabe75a084 SHA-256: 6440b188949e2eaf5594fa1f449c4bc8dc8cf97c28265b25ffc53ed473086fea
480 Risk Score

Malware Insights

MITRE ATT&CK
T1203 Exploitation for Client Execution T1059.007 JavaScript

The PDF file contains embedded JavaScript that exploits known Adobe Reader vulnerabilities, specifically CVE-2009-3459 and CVE-2007-5659, to achieve code execution. The JavaScript is designed to download a second-stage payload from the URL http://xxhsa.info/cgi-bin/smw/n002101801r0019R015ac620X90081768Ydc1bb891Z02f01010. The presence of multiple critical heuristic firings related to these CVEs and the embedded JavaScript confirms the exploitation and download mechanism.

Machine Learning

  • Nyx PDF Classifier malicious score 0.9999

Heuristics 15

  • Adobe Reader U3D auto-activated 3D annotation — CVE-2009-3459 critical CVE likely CVE_2009_3459_U3D_AUTOACTIVATE
    PDF contains a /Subtype /3D annotation that is configured to auto-activate on page view (/3DA <</A /PV /AIS /I>>) alongside a /U3D stream and JavaScript. This is the document shape used by CVE-2009-3459 (Adobe Reader U3D CLODProgressiveMeshDeclaration heap overflow, APSB09-15): the U3D parser runs without any user interaction once the page is rendered, while the accompanying JavaScript prepares a heap-spray to land controlled memory inside the corrupted allocation.
  • Collab.collectEmailInfo — CVE-2007-5659 critical CVE exact CVE_2007_5659
    PDF JavaScript calls Collab.collectEmailInfo — CVE-2007-5659 is a buffer overflow in Adobe Reader triggered by a long argument or heap-sprayed message field passed to Collab.collectEmailInfo(). Part of a series of Acrobat JS API exploits. (identified after JavaScript deobfuscation)
  • Adobe Reader U3D CLODProgressiveMeshDeclaration exploit critical CVE likely CVE_2009_3953
    PDF combines malformed U3D 3D content with JavaScript/action activation. CVE-2009-3953 is an Adobe Reader/Acrobat U3D CLODProgressiveMeshDeclaration array-boundary vulnerability triggered by malformed U3D data in a PDF.
  • U3D/3D content in PDF — Adobe Reader 3D parser CVE-family indicator high CVE related PDF_U3D_CVE_RELATED
    PDF contains U3D (Universal 3D) or 3D annotation content — CVE-2011-2462 and CVE-2009-3953 are critical vulnerabilities in Adobe Reader's U3D processing that allow arbitrary code execution. U3D content in PDFs is extremely rare in normal documents.
  • Adobe Reader APSB09-15 patch-range version gate (CVE-2009-3459) high CVE likely PDF_JS_ADOBE_APSB09_15_PATCH_GATE
    PDF JavaScript gates the exploit payload on an Adobe Reader version range that exactly matches the APSB09-15 patch boundary (< 9.2, < 8.1.7, < 7.1.4). That boundary fingerprints the CVE-2009-2990 / CVE-2009-3459 U3D parser vulnerabilities — no benign script tests all three Reader version-points together.
  • Adobe Reader APSB08-13 patch-range version gate (CVE-2007-5659) high CVE likely PDF_JS_ADOBE_APSB08_13_PATCH_GATE
    PDF JavaScript gates the exploit payload on (>= 8 && < 8.1.1) OR (< 7.1) — the Reader 7.0.x / 8.0–8.1.1 window patched by Adobe APSB08-13 for the CVE-2007-5659 Collab.collectEmailInfo buffer overflow. Only kits that target that exact bug check both of those patch points; benign scripts do not.
  • Generic recovered JavaScript exploit stage high PDF_GENERIC_STAGE_RECOVERY
    Bounded static stage recovery exposed hidden JavaScript through generic transforms such as null-byte collapse, percent decoding, marker replacement, arithmetic character codes, fromCharCode, numeric arrays, numeric-array minus-key decoders, alphabet-index arrays, /Producer half-difference metadata arrays, hex literals, marker-stripped Base64 literals, custom 6-bit XOR table decoders, or repeated-marker hex carriers. This rule is emitted only when the recovered stage contains exploit-like Acrobat JavaScript or shellcode markers.
  • Annotation subject callee-key hex JavaScript stager high PDF_ANNOT_SUBJECT_CALLEE_HEX_STAGER
    PDF JavaScript uses syncAnnotScan()/getAnnots() to read an indirect annotation /Subject stream, percent-decodes it through marker replacement, then uses a callee.toString()-derived key to decode and eval the final exploit stage.
  • PDF JavaScript shellcode contains an embedded download URL high PDF_JS_SHELLCODE_DOWNLOAD_URL
    Decoded PDF JavaScript shellcode contains a hardcoded http(s) URL stored as little-endian %uXXXX Unicode escapes. Reader exploit shellcode embeds the second-stage fetch URL this way and pulls it down with a urlmon/URLDownloadToFile-style download-and-execute (commodity downloader behaviour rather than a specific Acrobat CVE).
  • JavaScript action low PDF_JAVASCRIPT
    PDF contains a /JavaScript action. Generic JavaScript is common in benign forms; specific dangerous APIs are scored by separate rules.
  • Embedded JS stream low PDF_JS
    PDF references a /JS stream. Generic JavaScript is common in benign forms; specific dangerous APIs are scored by separate rules.
  • syncAnnotScan annotation-staging primitive low PDF_FOXIT_SYNCANNOTSCAN
    PDF JavaScript calls syncAnnotScan() — a no-op annotation-enumeration primitive used by exploit-kit JavaScript to stage payload reads from annotation /Subject fields before eval(). Not a vulnerable sink itself; rarely seen in legitimate PDFs. (matched in decompressed stream)
  • PDF differential parser failed info PDF_DIFFERENTIAL_PARSE_FAILED
    The cross-check parser (pdfminer.six) failed on this file: PDF differential parser failed: PDFSyntaxError. Static heuristics still ran and any of their findings above are valid; only the differential cross-check signal is missing.
  • Suspicious extracted artifact info EXTRACTED_FILE_STATIC_TRIAGE
    One or more files extracted from inside this sample matched static suspicious-content checks such as script obfuscation, encoded payload blobs, packed data, or execution/download terms.
  • Embedded URL info EMBEDDED_URL
    One or more URLs were extracted from the document. The URL itself is not a detection — see the per-URL labels for which channel (macro, JS, link annotation, document body, ...) reached each URL.
    URL http://xxhsa.info/cgi-bin/smw/n002101801r0019R015ac620X90081768Ydc1bb891Z02f01010

Extracted artifacts 6

Files carved from inside the sample during analysis.

FilenameKindSourceSize
javascript_obj0009_000.js
fc5b426b863fda3995fccd3f522ce82f4a2e780f06c61c18c1f9c560eb55c984
pdf-javascript-stream PDF /JS object 9 at offset 0x4692 493 bytes
generic_stage_recovery_000.js
fc660c3d76488f3ba6aec60bc2d814d849a416587e1ab5888082d6eceea73d74
deobfuscated-js generic stage recovery split-literal-normalize from JavaScript object 9 at offset 0x4692 488 bytes
annotation_subject_callee_hex_stage_000.js
8cea54e3f51acca322236835b30047c23a66111cdf9b5b382dbedca32f5e4973
deobfuscated-js annotation-subject callee-key decoded JavaScript at offset 0x1C17 5629 bytes
Detection
ClamAV: No threats found
Obfuscation or payload: likely
Carved artifact contains 5 eval/decoder/string-building token(s).
legacy_pdfkit_stage_000.js
d58377e5a1c4dc07ea4218c0b94a3536768f055c3f8baa8d6d84891e399cba82
deobfuscated-js repeated-marker hex decoded JavaScript at offset 0x1C6B 12716 bytes
Detection
ClamAV: No threats found
Obfuscation or payload: likely
Carved artifact contains 2 eval/decoder/string-building token(s). Carved artifact contains 1 long base64-like blob(s).
deobfuscated.js
ded9848c02ace0bace108845fabf4ce62eecddc2a9716aae09d90626b782621d
deobfuscated-js PDF JavaScript deobfuscation pass 242516 bytes
Detection
ClamAV: No threats found
Obfuscation or payload: likely
Carved artifact contains 2 long base64-like blob(s).
u3d_00_off000048db.bin
fb27c6f6ffe2b08a88f79f344ef3cb4f85460ac81a062194b6fd84ab6e48b628
pdf-3d-stream PDF U3D 3D stream at offset 0x48DB 28021 bytes
Detection
ClamAV: No threats found
Obfuscation or payload: likely
Carved artifact entropy is 7.79, consistent with packed or encrypted content.