Metraipolis › archive › dc_b2154ae14d80
TraceCompiler: Skill-Guided Mining and Compilation of LLM Agent Traces into Mostly Deterministic Workflows — arXiv:2608.02680
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2608.02680
- License (as archived)
- CC0-1.0
- Archived
- 2026-10-31T12:17:38.426Z
- SHA-256 of the archived text
fbc594d234dafca25fa617873e03041d722df5da449bec4c6cce25e862a10dda- Stable URL of this page
https://metraipolis.com/documento/dc_b2154ae14d80
How to cite
TraceCompiler: Skill-Guided Mining and Compilation of LLM Agent Traces into Mostly Deterministic Workflows — arXiv:2608.02680 — arXiv (en), https://arxiv.org/abs/2608.02680 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_b2154ae14d80 (Metraipolis archive dc_b2154ae14d80, sha256 fbc594d234dafca25fa617873e03041d722df5da449bec4c6cce25e862a10dda)
BibTeX
@misc{dcb2154ae14d80,
title = {TraceCompiler: Skill-Guided Mining and Compilation of LLM Agent Traces into Mostly Deterministic Workflows — arXiv:2608.02680},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2608.02680},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_b2154ae14d80; sha256 fbc594d234dafca25fa617873e03041d722df5da449bec4c6cce25e862a10dda},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2608.02680. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
Tool-using language-model agents repeatedly rediscover procedures they have already executed, producing traces that mix reusable structure with retries, exploration, accidental ordering, and repeated lookups. We present TraceCompiler, a skill-guided system that mines clusters of noisy agent traces and compiles them into executable, mostly deterministic workflows. It admits an inter-tool dependency only when a consumer argument contains a value attributable uniquely to an earlier producer; every hard edge carries an auditable evidence tuple, and ambiguous relations are marked suspected and impose no ordering constraint. Bindings are classified as constants, user inputs, copied outputs, transforms, or residual LLM decisions. On T1, a mechanized form of the rule recovers producer-consumer dependencies at 0.928 precision and 0.943 recall over 15,775 def-use edges of its training split, against 0.711 F1 for adjacency and 0.712 for a frequency-thresholded directly-follows measure on identical data; the compiler skill run blind reaches 0.992 on 250 of those edges. On AppWorld we replay released trajectories in the deterministic simulator to recover masked return values and measure the rule against 563 token edges at 0.993 precision - a self-consistency check, since replay injects tokens by a related heuristic. We compile two recurring intents: a Venmo money-request intent reduces 34 observed API calls to 11 runtime calls and, under leave-one-out execution against the benchmark's own state tests, passes 15 of 21, the failing fold escalating rather than acting because its required branch was never observed; and a Spotify/Todoist intent the compiler correctly refuses to compile, because an irreversible side effect is under-determined. We measure call reduction but not offline compilation cost, so we claim no net efficiency result. Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-08-05 in cs.AI). Categorie arXiv: Software Engineering (cs.SE) ; Artificial Intelligence (cs.AI); Machine Learning (cs.LG). Pagina del lavoro: https://arxiv.org/abs/2608.02680 — autori e testo completo sono lì (qui omessi i nomi: nessun dato personale nel deposito). Licenza: titolo e abstract sono metadati arXiv distribuiti in CC0 1.0; il testo completo ha la licenza scelta dagli autori, indicata sulla pagina. Letto il 2026-10-31T12:05:52.930Z dalla pagina https://arxiv.org/catchup/cs.AI/2026-08-05?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).