Metraipolis › archive › dc_8a590aa73d26

From Agent Behaviour to Agent-Friendly Documentation: An Empirical Study of How Coding Agents Discover, Read, and Write Technical Documentation — arXiv:2608.20195

Source
arXiv (en)
Original
https://arxiv.org/abs/2608.20195
License (as archived)
CC0-1.0
Archived
2026-10-18T11:49:51.423Z
SHA-256 of the archived text
e2418e21425f46069a9e32cca5b56057a47e84bc6d7dc0ab6367093251931b0e
Stable URL of this page
https://metraipolis.com/documento/dc_8a590aa73d26

How to cite

From Agent Behaviour to Agent-Friendly Documentation: An Empirical Study of How Coding Agents Discover, Read, and Write Technical Documentation — arXiv:2608.20195 — arXiv (en), https://arxiv.org/abs/2608.20195 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_8a590aa73d26 (Metraipolis archive dc_8a590aa73d26, sha256 e2418e21425f46069a9e32cca5b56057a47e84bc6d7dc0ab6367093251931b0e)
BibTeX
@misc{dc8a590aa73d26,
  title = {From Agent Behaviour to Agent-Friendly Documentation: An Empirical Study of How Coding Agents Discover, Read, and Write Technical Documentation — arXiv:2608.20195},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2608.20195},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_8a590aa73d26; sha256 e2418e21425f46069a9e32cca5b56057a47e84bc6d7dc0ab6367093251931b0e},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2608.20195. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Technical documentation is written for human developers, but an increasing share of software changes is now authored by autonomous coding agents. Which documents they consult, when, and what follows remain unknown. We conduct a behaviour-grounded study of agent-documentation interaction across two public datasets: 557 agentic coding sessions from SWE-chat, yielding 94,813 development events including 3,033 documentation interactions; and 33,097 agentic pull requests from AIDev, with 690,260 classified file-level change records. Four findings challenge current documentation practice. First, agents' documentation work is dominated by agent-facing artefacts: instruction files and working notes account for 60.5% of all documentation interactions, versus 10.6% for classical technical documentation and 1.3% for API references. Second, the link between consultation and code editing is unresolved: the adjacent transition probability is 0.002 and the unadjusted three-event lift 1.05, whereas a stage-adjusted model places it above unity (OR 1.33 [1.09, 1.62]); documentation creation is elevated unadjusted (lift 1.67) but its adjusted interval includes unity. Third, no explicit documentation-based validation sequence was observed, and consultation is associated with less immediate testing (lift 0.23, cluster CI 0.08-0.45; adjusted OR 0.39 [0.25, 0.60]). Fourth, consultation is self-initiated (70.2%) far more often than failure-driven (7.5%), and documentation trails code: among multi-commit pull requests changing both, code is touched first 4.7x more often. From these traces we derive a descriptive model of agent-documentation interaction as a two-lobed cycle rather than a linear journey, and show that two widely assumed properties of "agent-friendly" documentation - actionability and verifiability - lack consistent behavioural support. We release our pipeline, coding scheme, and event-level data.

Tema degli esploratori: calcolo e infrastrutture (retro-catalogo arXiv: annunci del 2026-08-21 in cs.SE).
Categorie arXiv: Software Engineering (cs.SE) ; Artificial Intelligence (cs.AI); Human-Computer Interaction (cs.HC).
Pagina del lavoro: https://arxiv.org/abs/2608.20195 — 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 2026-10-18T11:48:10.195Z dalla pagina https://arxiv.org/catchup/cs.SE/2026-08-21?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).