pith:C43N4527
Content-Aware Attack Detection in LLM Agent Tool-Call Traffic: An Empirical Study of Features, Architectures, and Evaluation Protocols
MCPShield shows that content embeddings of tool arguments and responses are essential for detecting attacks on LLM agent traffic, lifting AUROC from 0.64 to above 0.89.
arxiv:2605.11053 v3 · 2026-05-11 · cs.CR · cs.AI · cs.LG
Add to your LaTeX paper
\usepackage{pith}
\pithnumber{C43N4527I7EBNA4U3P6FEMSGNT}
Prints a linked badge after your title and injects PDF metadata. Compiles on arXiv. Learn more · Embed verified badge
Record completeness
Claims
Content-level features are essential: metadata-only detection plateaus around an AUROC of 0.64 regardless of architecture, while content embeddings push the AUROC above 0.89.
The attack examples in RAS-Eval and ATBench are representative of real-world threats and that SBERT embeddings capture generalizable attack signals rather than dataset-specific artifacts.
MCPShield detects attacks on LLM agent tool-call traffic by encoding sessions as graphs enriched with SBERT content embeddings, achieving AUROC above 0.89 with content features versus 0.64 for metadata alone.
Formal links
Receipt and verification
| First computed | 2026-05-25T02:01:22.999408Z |
|---|---|
| Builder | pith-number-builder-2026-05-17-v1 |
| Signature | Pith Ed25519
(pith-v1-2026-05) · public key |
| Schema | pith-number/v1.0 |
Canonical hash
1736de775f47c8168394dbfc5232466cc48c2b225a6cac2e82cdcf2700ae03c2
Aliases
· · · · ·Agent API
Verify this Pith Number yourself
curl -sH 'Accept: application/ld+json' https://pith.science/pith/C43N4527I7EBNA4U3P6FEMSGNT \
| jq -c '.canonical_record' \
| python3 -c "import sys,json,hashlib; b=json.dumps(json.loads(sys.stdin.read()), sort_keys=True, separators=(',',':'), ensure_ascii=False).encode(); print(hashlib.sha256(b).hexdigest())"
# expect: 1736de775f47c8168394dbfc5232466cc48c2b225a6cac2e82cdcf2700ae03c2
Canonical record JSON
{
"metadata": {
"abstract_canon_sha256": "93766ad117e27044b395a3531afd639c203eda23c05c7c1bc349e4308001ddeb",
"cross_cats_sorted": [
"cs.AI",
"cs.LG"
],
"license": "http://arxiv.org/licenses/nonexclusive-distrib/1.0/",
"primary_cat": "cs.CR",
"submitted_at": "2026-05-11T14:55:48Z",
"title_canon_sha256": "56536732f9de1996a11d36b08864f8e9afceabd57fc31b3cd927a753d8126632"
},
"schema_version": "1.0",
"source": {
"id": "2605.11053",
"kind": "arxiv",
"version": 3
}
}