# AI Guardrail Survival under Single-Cycle Agentic Self-Summarization

arXiv: https://arxiv.org/abs/2608.11392 · HTML: https://mlsanity.com/paper/2608.11392/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Placed in Agent threats, guardrails and governance, one of 16 directions in LLM agents and tool use.
- Builds on 6 of its 8 references that are in the map.
- First posted on arXiv 11 Aug 2026.

## Placed in

- LLM agents and tool use / [Agent threats, guardrails and governance](https://mlsanity.com/agents/agent-threats-guardrails-and-governance/)
- LLM safety and security / [Prompt injection and agent security](https://mlsanity.com/safety/prompt-injection-and-agent-security/)

## Builds on (most cited first, 6 of 6)

- [ACON: Optimizing Context Compression for Long-horizon LLM Agents](https://mlsanity.com/paper/2510.00615/) (2025)
- [Enforcing Temporal Constraints for LLM Agents](https://mlsanity.com/paper/2512.23738/) (2025)
- [Noisy but Valid: Robust Statistical Evaluation of LLMs with Imperfect Judges](https://mlsanity.com/paper/2601.20913/) (2026)
- [When Refusals Fail: Unstable Safety Mechanisms in Long-Context LLM Agents](https://mlsanity.com/paper/2512.02445/) (2025)
- [Evaluating the Sensitivity of LLMs to Prior Context](https://mlsanity.com/paper/2506.00069/) (2025)
- [Omission Constraints Decay While Commission Constraints Persist in Long-Context LLM Agents](https://mlsanity.com/paper/2604.20911/) (2026)

## Abstract (arXiv)

Long-running agents periodically compact their context, replacing the transcript with a model-generated summary. Recent work shows that dropping a standing safety constraint during compaction drives behavioral violations across many models (Governance Decay; Chen, 2026). We ask a finer question: under a single compaction cycle, how is a safety rule lost, and what does that imply for detection and evaluation? Our central finding is that a presence check is not a safety check: when compaction does not drop a rule outright, it often leaves something that looks like a rule but does not act like one. On behavioral replay, a degraded residue leads the model to perform the prohibited action far more often than an intact welded rule does (all-case gaps of +34 and +57 points under two replay models, both positive), category-level survival behaves like a residue, and even intact rules sometimes fail to fire, so an audit that checks only textual presence gives false assurance. Sharpening this, rule-form items are retained substantially more often than prominence-matched facts, which is exactly why presence-based checking feels adequate even though survival is not protection. Textual loss is regime-dependent (weld-or-drop with a single rule; degraded predicate-loss residues under a tighter budget), and we did not observe the hypothesized textual severing mode. Such loss is silent at runtime and detectable only by comparison with retained external ground truth (such as a constraint registry), which reveals textual absence but not whether a surviving rule still fires. We also document evaluation pitfalls where LLM-judge labels alone would have reversed a conclusion. All results concern a single compaction cycle.