Datasets:

TonyDzi
/

agent-sessions

session_id stringlengths 10 10	date stringdate 2026-08-24 00:00:00 2026-10-29 00:00:00	words int32 6 263k	turns int32 1 1.5k	tools_used listlengths 0 44	models listlengths 0 3	effort listlengths 0 3	text stringlengths 39 2.16M	extra stringclasses 4 values
s_154e2276	2026-08-24	7,508	19	[ "Bash", "Edit", "Grep", "PowerShell", "Read", "Skill", "Write" ]	[ "claude-fable-5" ]	[ "medium" ]	User: изучи этот комп и скажи какую рабочую папку мне стоит выбирать при старте новой сессии --- Claude: [⏰ понедельник, утро, 08:30, 24 августа 2026] Сейчас быстро осмотрю машину: какие папки-кандидаты есть, где живут прошлые сессии и память, и дам вердикт одной строкой. --- > `PowerShell` -- {"command"...	{"class": "mixed"}
s_1aa99262	2026-08-24	3,488	13	[ "Bash", "Edit", "SendUserFile", "TaskOutput", "ToolSearch", "WebSearch", "Write", "mcp__ccd_session_mgmt__set_session_title", "mcp__claude-in-chrome__list_connected_browsers" ]	[ "claude-opus-5" ]	[ "high" ]	User: <scheduled-task name="auto-hub-260824-NODE-0X-folder-cleanup" file="C:\Users\dev\.claude\scheduled-tasks\auto-hub-260824-NODE-0X-folder-cleanup\SKILL.md"> This is an automated run of a scheduled task. The user is not present to answer questions. For implementation details, execute autonomously without asking...	{"class": "hold"}
s_1b42f78a	2026-08-24	43,205	69	["Bash","Edit","Grep","PowerShell","Read","Skill","ToolSearch","Write","mcp__scheduled-tasks__create(...TRUNCATED)	[ "claude-fable-5" ]	[ "medium" ]	"User:\n\nизучи этот комп и скажи какую рабочую папку мне(...TRUNCATED)	{"class": "mixed"}
s_3ee851b6	2026-08-24	205	2	[ "Bash", "ToolSearch", "mcp__ccd_session_mgmt__set_session_title" ]	[ "claude-opus-5" ]	[ "high" ]	"User:\n\n<scheduled-task name=\"auto-hub-260824-NODE-0X-folder-cleanup\" file=\"C:\\Users\\dev\(...TRUNCATED)	{"class": "unknown"}
s_4ff9f99f	2026-08-24	1,383	7	["Bash","ToolSearch","mcp__scheduled-tasks__list_scheduled_tasks","mcp__scheduled-tasks__update_sche(...TRUNCATED)	[ "claude-opus-5" ]	[ "high" ]	"User:\n\n<scheduled-task name=\"voice-dispatcher\" file=\"C:\\Users\\dev\\.claude\\scheduled-ta(...TRUNCATED)	{"class": "publish"}
s_65d8e7b9	2026-08-24	6,370	50	[ "Bash", "Edit", "PowerShell", "ToolSearch", "Write", "mcp__claude-in-chrome__list_connected_browsers" ]	[ "claude-opus-5" ]	[ "high" ]	"User:\n\n<scheduled-task name=\"auto-hub-260824-dr-runner-hiring\" file=\"C:\\Users\\dev\\.clau(...TRUNCATED)	{"class": "mixed"}
s_72759f6a	2026-08-24	2,097	6	[ "Bash", "SendUserFile" ]	[ "claude-opus-5" ]	[ "high" ]	"User:\n\n<scheduled-task name=\"auto-hub-260824-nsfw-dr-posts\" file=\"C:\\Users\\dev\\.claude\(...TRUNCATED)	{"class": "publish"}
s_cd326caf	2026-08-24	3,104	33	[ "Bash", "PowerShell", "SendUserFile" ]	[ "claude-opus-5" ]	[ "high" ]	"User:\n\n<scheduled-task name=\"auto-hub-260824-NODE-0X-folder-cleanup\" file=\"C:\\Users\\dev\(...TRUNCATED)	{"class": "publish"}
s_f6ad222d	2026-08-24	1,000	2	[ "Agent", "Bash" ]	[ "claude-opus-5" ]	[ "high" ]	"User:\n\n<scheduled-task name=\"content-miner-nightly\" file=\"C:\\Users\\dev\\.claude\\schedul(...TRUNCATED)	{"class": "mixed"}
s_05a0ad6f	2026-08-25	999	3	[ "Bash", "PowerShell" ]	[ "claude-opus-5" ]	[ "high" ]	"User:\n\n<scheduled-task name=\"task-hygiene-shadow\" file=\"C:\\Users\\dev\\.claude\\scheduled(...TRUNCATED)	{"class": "publish"}

End of preview. Expand in Data Studio

Agent Sessions: real production transcripts of an LLM coding agent

Thousands of blog posts describe AI agents; very few open corpora show real, unedited working sessions where an agent builds, breaks and repairs things for months on someone's live infrastructure. This is one. Every row is one complete session between a human operator and an LLM coding agent (Claude Code) running on a small fleet of personal machines. Not demos, not curated examples. The agent is wrong a lot, and the record keeps that in.

The dataset is revised daily. Each revision is one commit tagged vYYYY.MM.DD; pin a tag or a commit sha to reproduce a result. Archived snapshots with a DOI: 10.5281/zenodo.23712140. Readable mirror: https://github.com/tonydzi/agent-sessions.

This revision (2026-11-05)

Numbers below are printed by the release script from the data it uploaded, not typed by hand.

12 558 sessions, 30 457 333 words
date range: 2026-08-24 .. 2026-10-29
most frequent tools: Bash, Read, ToolSearch, Write, Edit, PowerShell
held back by the secret/PII gate at upload: 0

Sessions by day

Newest first. Every session id of every day is listed in sessions_by_day.csv; the Dataset Viewer (parquet shards) can sort or filter by the date column to read one day.

day	sessions	words
2026-10-29	40	56 735
2026-10-28	238	1 084 312
2026-10-27	162	553 525
2026-10-26	190	350 142
2026-10-25	235	581 409
2026-10-24	168	786 207
2026-10-23	227	1 675 511
2026-10-22	152	795 417
2026-10-21	187	874 211
2026-10-20	197	747 097
2026-10-19	213	1 009 920
2026-10-18	136	436 414
2026-10-17	131	311 638
2026-10-16	164	643 871
2026-10-15	182	378 853
2026-10-14	263	1 526 212
2026-10-13	160	333 149
2026-10-12	203	405 385
2026-10-11	196	315 166
2026-10-10	240	184 619
2026-10-09	272	474 512
2026-10-08	223	677 877
2026-10-07	265	1 085 696
2026-10-06	337	2 480 463
2026-10-05	134	264 411
2026-10-04	125	187 557
2026-10-03	180	279 817
2026-10-02	260	468 450
2026-10-01	257	873 315
2026-09-30	269	653 538
2026-09-29	249	297 409
2026-09-28	245	538 787
2026-09-27	219	284 648
2026-09-26	292	168 265
2026-09-25	239	200 861
2026-09-24	556	288 906
2026-09-23	489	1 041 421
2026-09-22	489	500 079
2026-09-21	456	649 024
2026-09-20	560	475 423
2026-09-19	782	198 862
2026-09-18	819	215 236
2026-09-17	216	187 317
2026-09-16	20	116 513
2026-09-15	67	552 189
2026-09-14	55	274 479
2026-09-13	49	184 910
2026-09-12	34	44 665
2026-09-11	55	336 677
2026-09-10	79	545 125
2026-09-09	44	263 526
2026-09-08	40	101 479
2026-09-07	35	129 489
2026-09-06	38	99 715
2026-09-05	50	278 997
2026-09-04	47	498 035
2026-09-03	51	348 045
2026-09-02	51	183 876
2026-09-01	46	432 709
2026-08-31	31	52 216
2026-08-30	45	166 276
2026-08-29	40	86 751
2026-08-27	4	19 328
2026-08-26	25	52 409
2026-08-25	26	79 897
2026-08-24	9	68 360

Selection

Since 1 Nov 2026 every mature session is released verbatim: work, hiring, money, leads and family included. Topic is no longer a filter. Only two things keep a session out: a secret that cannot be cut, or a personal-data finding that cannot be masked. Such rows are held back and counted, never silently dropped.

Builder counters for this release (dataset_meta.json): rejected before release: blurred 43. A session enters only after it is 7 days old, where age is the last timestamp inside the transcript (file mtime lies: reapers and continuations touch files).

Files

file	rows
`data/2026-08-00.parquet`	180
`data/2026-09-00.parquet`	6 641
`data/2026-10-00.parquet`	5 737

data/*.parquet: zstd shards, one or more per month (YYYY-MM-NN). This is what load_dataset("TonyDzi/agent-sessions") and the Dataset Viewer read.
agent_sessions.jsonl: the same rows as one JSON-lines file, for grep and jq.
sessions_by_day.csv: date,session_id,words,turns, one line per session, newest day first.
dataset_meta.json: exact build counters, rejection reasons, shard list.
agent_sessions.parquet at the repository root is a frozen snapshot from 30 Oct 2026, kept for old links; it is not part of the default config.

field	type	meaning
`session_id`	string	stable pseudonymous id, `s_` + 8 hex; not the real session uuid
`date`	string	`YYYY-MM-DD` of the last timestamp in the transcript
`words`	int32	word count of `text`
`turns`	int32	human + assistant text turns
`tools_used`	list[string]	distinct tools the agent called
`models`	list[string]	model ids seen in the session
`effort`	list[string]	reasoning-effort settings seen
`text`	string	rendered session: turns, tool calls and their arguments
`extra`	string	JSON of any further builder fields, else null

Language: mostly Russian prose with English technical vocabulary, shell and code.

How it was rendered

The raw material is the agent harness's own .jsonl transcripts. About 97% of a raw file is machinery (tool results, screenshots, system reminders). Rendering keeps the human turns, the assistant turns and tool calls with arguments; the speaker label is a role (**User:**), not a name. The pipeline is deterministic code; an LLM reviewer only flags meaning-level personal data.

Masking (replacement, not deletion)

A reader needs the shape of a path or an identifier to follow the agent, so values are replaced with stable placeholders instead of vanishing.

Secrets (private keys, API/token shapes, password=/token= assignments, chat ids) are cut. A session with a secret that cannot be cut is not released: the gate is fail-closed and runs twice, once in the builder and once again in this upload script.
Paths and machines: home directories become C:\Users\dev, vault paths a neutral name, hosts HUB-01 / NODE-0X, including the harness's dash-encoded path form.
Session uuids and long numeric ids map to stable s_… / id_… tokens, so references between sessions resolve inside the corpus and nowhere else.
People (mask rule v2, 30 Oct 2026): a first name becomes a stable pseudonym, the same in every session, so recurring people stay recurring. No first letters survive: surnames become [ФАМИЛИЯ], e-mail [EMAIL], phone [ТЕЛЕФОН], street address / flat / postal code [АДРЕС], age [ВОЗРАСТ]. Public researchers cited in passing are masked too.
Quasi-identifiers (third-party employer, client, city, rare details) are not regex-shaped: an independent LLM reviewer flags them and the finding is re-masked ([ИМЯ] / [СКРЫТО]). The reviewer's notes quote what they found, so they are never published.

What is not hidden: the operator publishes this corpus under his own name on purpose. His projects, repositories, methods and conclusions are the subject of the dataset.

If you find something that should have been masked, open a discussion on this repository; the row is re-masked in the next daily revision.

License

CC BY 4.0. Attribute as: Anton D., Agent Sessions: real production transcripts of an LLM coding agent, Hugging Face, 2026, doi:10.5281/zenodo.23712140. Tool output quoted inside sessions keeps its upstream license.

Limitations

One operator, one agent family, one fleet: this is a deep sample, not a broad one. Masking replaces tokens and so can break the grammar of a sentence. Sessions are rendered, not raw: tool results are dropped, so an agent's claim can be checked only against its next actions.

Downloads last month: 195