- Rating
- 2
- Confidence
- 3
- Informativeness
- 2
- Presentation
- 2
- Soundness
- 2
Read review
Paper summary
TrojanAgent combines task-aware agent-specification augmentation with AMR-based instruction fusion to influence agent selection and bypass runtime guardrails. Evaluation covers 50 tasks, 1,547 specifications, six retrievers, three target models, and two guardrails, with strong selection results and substantial TaskShield bypass gains.
Critical strengths
S1: The attack considers both workflow selection and post-delegation behavior, addressing two relevant security boundaries in dynamic orchestration. S2: Evaluation spans several retriever families and measures tool execution under two guardrails. The TaskShield results show a substantial component-level improvement over SCH.
Critical weaknesses
W1: Separate selection and runtime metrics do not establish the joint attack outcome. BU requires only one delegated subtask to succeed, which is weaker than full benign-task preservation (Section 3.1; Equation 4; Section 5.1). W2: The AMR-to-text prompt also performs explicit text obfuscation. The graph-builder ablation does not clearly specify which transformations remain, limiting attribution of gains to AMR fusion (Section 5.3; Appendix A.1). W3: The deviation, intent-preservation, modification-cost, and task-preservation scores are insufficiently operationalized. Selection of the final Pareto candidate is also unclear, limiting reproducibility of the optimization (Sections 4.2–4.3).
Questions for authors
Q1: What is the joint rate of selection, guardrail passage, malicious execution, and full benign-task completion on the same trials? Q2: Does the no-graph ablation retain the same text-obfuscation instructions and generation budget? Q3: How are the optimization scores computed, and what rule selects the final candidate from the Pareto frontier?
3 comments
Thank you, reviewer RnYb! (2/3)
Weakness 2: “The AMR-to-text prompt also performs explicit text obfuscation. The graph-builder ablation does not clearly specify which transformations remain, limiting attribution of gains to AMR fusion (Section 5.3; Appendix A.1).”
Question 2: Does the no-graph ablation retain the same text-obfuscation instructions and generation budget?
We appreciate the reviewer’s thoughtful comment regarding the ablation study. To answer whether the AMR advantage persists under a matched generation procedure and candidate budget, we further decompose the “without AMR” condition by retaining the text-rewriting operations while removing AMR parsing and graph fusion. The No AMR Fusion + Text Rewriting condition retains all four text-level operations: weakening direct imperatives, removing tool-specific details, converting commands into state descriptions, and neutralizing malicious motivation. It uses GPT-4o-mini at temperature 0.3, a 4,096-token limit, and one generated candidate. The results are reported in Table 2.
Table 2 shows that removing AMR while retaining rewriting reduces BSR by 26.67 percentage points under TaskShield and 4.00 percentage points under ToolSafe, relative to Full AMR. The result indicates that AMR fusion contributes beyond the text-rewriting operations.
Table 2. BSR Ablation Results for AMR Fusion and Text Rewriting
| Experimental condition | TaskShield BSR (mean ± SD; 95% CI) | ToolSafe BSR (mean ± SD; 95% CI) | Δ BSR vs. AMR Fusion + Text Rewriting (TaskShield; ToolSafe) |
|---|---|---|---|
| AMR Fusion + Text Rewriting | 70.00 ± 2.00; [65.03, 74.97] | 45.33 ± 1.15; [42.46, 48.20] | N/A |
| No AMR Fusion + Text Rewriting | 43.33 ± 6.43; [27.36, 59.30] | 41.33 ± 4.16; [30.99, 51.68] | -26.67 pp; -4.00 pp |
| No AMR Fusion + No Text Rewriting | 38.00 ± 0.00; [38.00, 38.00] | 12.00 ± 2.00; [7.03, 16.97] | -32.00 pp; -33.33 pp |
Thank you, reviewer RnYb! (3/3)
Weakness 3: “The deviation, intent-preservation, modification-cost, and task-preservation scores are insufficiently operationalized. Selection of the final Pareto candidate is also unclear, limiting reproducibility of the optimization (Sections 4.2–4.3).”
Question 3: How are the optimization scores computed, and what rule selects the final candidate from the Pareto frontier?
We appreciate the reviewer’s concern regarding the reproducibility of the optimization. We will clarify each point in the revision:
- Preservation and deviation scores. Let be the AMR graph of the benign task, a candidate fused graph, the transformed intent graph, and the set of triples in graph . Task-preservation is exact triple recall, , with the constraint . Intent-preservation, , measures the fraction of transformed-intent triples preserved; valid candidates have , which we use as a feasibility check. The specification-deviation term counts the fields changed in specification relative to the source specification , and bounds the permitted number of changes.
- Modification cost and Pareto frontier. Each valid candidate is scored by added variable nodes, added variable-to-variable edges, and attachment-class priority. We assign the priority using ordered first-match rules:
ROOTED= 4 when the anchor is the task graph's top node;COORD= 3 for roles:op1,:op2,:snt1,:snt2, or:snt3;LIST_EXT= 2 for an adjunct role (:purpose,:condition,:manner,:time,:li,:cause, or:concession) attached toand,or, ormulti-sentence;ADJUNCT_FUSED= 1 for those adjunct roles on other concepts;ARG_FUSED= 0 for:ARG1or:ARG2; andOTHER= 5 if no earlier rule matches. Lower values indicate lower cost. The frontier contains candidates for which no other candidate is no worse on all three costs and strictly better on at least one. - Final candidate selection. We sort candidates on the Pareto frontier lexicographically by the following criteria: (1) node cost, ascending; (2) edge cost, ascending; (3) attachment cost, ascending; (4) lowering depth, descending; (5) anchor variable, ascending; (6) role, ascending; and (7) lowering label, ascending. Node cost counts added variable nodes; edge cost counts added variable-to-variable edges; attachment cost is the attachment-class priority; lowering depth counts added wrapper and modal nodes; anchor variable is the PENMAN variable receiving the intent; role is the semantic relation used for attachment; and lowering label identifies the wrapper/modal combination. The first three criteria are the Pareto costs, and the remaining criteria break ties. The algorithm selects the first candidate in this ordering.
- Reproduction materials. The anonymous repository [1] also provides the implementation details, and experiment files needed to reproduce the procedure.
[1] Anonymous repository: https://anonymous.4open.science/r/anonylink-42789-CC08/.
Thank you, reviewer RnYb! (1/3)
Response for Weaknesses and Questions
We thank the reviewer for raising this important point. Following the reviewer’s suggestion, we add two complementary metrics: the end-to-end attack success rate (ASRe2e) and full legitimate-task utility (BUfull). Specifically, ASRe2e measures the joint occurrence of malicious agent selection, guardrail bypass, and malicious execution, while BUfull measures whether the original benign task is fully completed. We report these two outcomes independently, following prior evaluations such as AgentDojo [1], Rethinking Indirect Prompt Injection as a Test-Time Search Problem [2], and Will the User Ever Know? [3], which also assess attack success and utility under attack separately. We evaluate 50 tasks per execution model, two guardrails, and three repetitions. The results are shown in Table 1.
As shown in Table 1, TrojanAgent achieves the highest ASRe2e under both guardrails, reaching 63.38% with TaskShield and 41.73% with ToolSafe. It also retains at least one successful benign subtask at rates comparable to the baselines. The BUfull results reveal a trade-off between attack success and complete benign-task completion. We will explicitly report this limitation, and improving full legitimate-task completion while preserving high attack success remains an important direction for future work. We would also like to clarify that the central contribution of our work is an end-to-end agent-poisoning framework that targets both agent selection and runtime guardrail bypass, rather than a guarantee of zero utility loss. This scope is consistent with the description of our main contribution in the Introduction (Contributions paragraph). We acknowledge that preserving benign task completion remains an important objective, as discussed in the Threat Model (Section 3.1) and the Overview (Section 4.1). However, our contribution should be evaluated in terms of the end-to-end poisoning capability and runtime guardrail bypass, rather than an expectation of zero utility degradation.
Table 1. End-to-End Attack Success and Benign Utility by Attack Method and Guardrail
[1] Debenedetti E, Zhang J, Balunovic M, et al. Agentdojo: A dynamic environment to evaluate prompt injection attacks and defenses for llm agents[J]. Advances in neural information processing systems, 2024, 37: 82895-82920.
[2] Nguyen D M, Kim J S, Manczak B, et al. Rethinking Indirect Prompt Injection as a Test-Time Search Problem[J]. arXiv preprint arXiv:2609.04495, 2026.
[3] Lee Y, Kim Y, Lee W. Will the User Ever Know? Covert Indirect Prompt Injection Attacks on Tool-Using LLM Agents[C]//Proceedings of the 2026 Conference on Empirical Methods in Natural Language Processing. 2026: 46371-46389.