---
title: "Jiro vs Claude Tag (2026): Billing, Models, Control and Slack Compared"
canonical: https://jiro.bot/compare/claude-tag
last_updated: 2026-11-06
---

# Jiro vs Claude Tag (2026): Billing, Models, Control and Slack Compared

Updated 2026-11-06

> **Should my team use Jiro or Claude Tag?**
>
> Pick Jiro if you run a small engineering team, five to twenty people, whose Claude or ChatGPT seats are already paid, and you want a flat $99 or $250 a month, a choice of Claude Code, Codex or Cursor Agent, and control over the environment. Pick Claude Tag if you are on Claude Team or Enterprise, want Anthropic to run everything, and accept usage billing from an organization balance. Jiro is a background coding agent by Nori. It is not Amazon's Kiro or Atlassian's Jira.

## Jiro vs Claude Tag, twelve rows

|  | Jiro | Claude Tag |
|---|---|---|
| Plan you need | Your Claude/ChatGPT login or API key; Cursor key [1] | Team and Enterprise only [2] |
| Price | Free 30-day trial; Developer $99/mo; Team $250/mo [3] | In Team plan: $20/seat/mo annual, $25 monthly [4] |
| How usage is billed | Your keys, zero token markup [5] | Usage-billed, no per-seat charge [6] |
| Agents and models | Claude Code, Codex, Cursor, OpenCode, pi-acp [7] | Claude models only [8] |
| Where a session runs | Isolated cloud environment with your repos and tools [9] | Anthropic cloud sandbox per thread [10] |
| Environment setup | Org setup script prepares each environment [11] | Admin setup script per environment; CLAUDE.md installs [12] |
| Repo hooks, MCP, effort | MCP server catalog in the dashboard [13] | No repo hooks, .mcp.json or effort setting [14] |
| Slack behaviour | Mention in Slack; works in the thread [15] | Shared @Claude teammate in Slack channels [16] |
| Schedules and events | Cron, one-off reminders, webhooks [17] | Schedules; no repo-event triggers [18] |
| What comes back | PRs, commits, comments for review [19] | Draft PRs via Claude GitHub App; GitLab token [20] |
| Inside your own infrastructure | Nori cloud, or a dedicated deployment in your AWS account [21] | Self-hosted runners, without connectors [22] |
| Product status | Paid plans after a free 30-day trial | Public beta [23] |

Claude Tag cells come from Anthropic's documentation; Jiro cells from Nori's site and docs. All checked 2026-11-06.

## What each one costs, with units

|  | Jiro | Claude Tag |
|---|---|---|
| Smallest paid plan | Developer, $99 a month, one user, up to three runtimes | Team standard seat, $20 a month annual or $25 monthly, 2 to 150 seats |
| Larger team | Team, $250 a month, five shared runtimes | Enterprise: $20/seat/mo annual, 20-seat minimum, plus usage [24] |
| Model tokens | Your own Claude or ChatGPT plan or API key, no markup | Organization usage balance, funded by an Owner |
| Prepayment | None beyond the plan | Team plan must prepay credits [25] |
| When the budget runs out | Your subscription's own limits; extra usage at API rates | At the cap, Claude stops mid-task [26] |
| Trial | 30 days, no card | No published per-task cost; pilot with a cap [27] |

Jiro figures are from Nori's pricing page; Claude Tag's from Anthropic's pricing and spend-limit pages, cited in the prose below.

## What is the short answer?

Both tools put a coding agent in Slack and both open pull requests. They differ on who pays for tokens, who picks the model, and who controls the sandbox. Claude Tag is Anthropic's own product: Opus or Sonnet, billed by usage from an organization balance, running on infrastructure Anthropic configures [6] [10]. Jiro is a layer on top of agents you already license: Claude Code, Codex or Cursor Agent, driven by a login or key you already own, for a fixed monthly price [1] [3]. A Claude Enterprise organization that wants zero decisions should take Tag. A dozen-person team already funding Claude or ChatGPT seats, and wary of a second usage bill, should take Jiro. Everything below is the evidence for those two sentences.

## How does Jiro bill compared with Claude Tag?

Jiro is a flat line item. One user and up to three runtimes cost $99 a month on Developer; several users sharing five runtimes cost $250 a month on Team; before either, a trial runs 30 days and asks for no card [3]. Runtimes that sit idle sleep, waking when called [28]. Model tokens come from the Claude or ChatGPT plan you already hold, and Nori charges zero markup on them [5]. The variable part therefore lives on a bill you already manage: Max costs $100 monthly and allows five times what Pro does per five-hour window and per week, with the heavier tier now branded Ultra [29], and usage past those limits continues as credits at API rates [30].

Claude Tag adds no per-seat charge. Channel and group-DM work draws by usage from a balance an Owner funds, while a one-to-one DM bills to the sender's own seat [6]. A Team organization must buy credits before Claude answers in channels [25]. Owners set a monthly cap and per-channel caps, and Claude warns in Slack as usage nears them [31]; at the cap, it stops mid-task [26]. Anthropic publishes no per-task price and recommends a capped pilot [27]. Predictability favours Jiro. Fewer vendors favours Tag.

## Who is allowed to buy each one?

Claude Tag is available on Team and Enterprise plans only, not on Free, Pro, Max or Ultra, and not through third-party deployments [2]. Those Team seats are priced at $25 each per month, or $20 when billed yearly, and the plan spans 2 to 150 people [4]; Enterprise starts at 20 seats, $20 apiece on annual terms, and bills usage separately at API rates [24]. Organizations with Zero Data Retention enabled cannot use it at all [32].

Jiro has no plan gate. Any Claude subscriber, individual or organizational, runs `claude setup-token` to mint a token that lasts a year [33] and pastes it in; for Codex the login is ChatGPT OAuth or an OpenAI key, and Cursor Agent requires Cursor's own key [1]. That difference alone decides the question for many startups. If your engineers are on Max, Tag is not an option today, and Jiro is.

## Which models and agents can each run?

Claude Tag runs Anthropic's Opus and Sonnet models, chosen per thread, per channel or by admin default [8]. For a team that has settled on Claude for the next two years, that restriction costs nothing. Fast mode on Opus is available once an Owner allows it, and the reply footer names the model that answered.

Jiro's first-class harnesses are five: Claude Code and Codex, Cursor Agent, and the open OpenCode and pi-acp [7], so one channel can use Claude for most work and hand a stubborn refactor to Codex without a procurement cycle. The models available inside each harness follow whatever your login or key unlocks. The harness also differs. Tag skips repository hooks and `.mcp.json`, has no effort setting, and runs in auto mode where Claude's permission checker reviews each action and admins pre-approve routine ones with allow rules [14] [34]. Jiro runs Claude Code through the Agent SDK bundle, with an MCP catalog in the dashboard where each server's own permissions still apply [13].

## How much control do you get over the environment?

Tag gives admins real levers, just fewer of them. An admin can add install commands to the setup script of the environment a channel's sessions run on, or put per-repository installs in CLAUDE.md, which repeat every session because each sandbox is fresh [12]. After a granted repository is cloned, its `CLAUDE.md`, rules files and committed skills take effect, but `.mcp.json` stays unread and `.claude/settings.json` hooks stay inert [35]. Connectors, plugins, skills and remote MCP servers can be added by an admin; local stdio MCP servers cannot [36]. Outbound traffic is default-deny except GitHub, limited to hosts an allow layer covers [37].

Jiro's model is one isolated cloud environment per session, arriving with the repository checked out and the dependencies, tools and services already in place [9], prepared by an organization setup script [11]. One claimed machine serves one active session [38]. Shared context is authored once as a skillset; Nori then emits the file each harness reads, so nothing is maintained twice [39]. Control favours Jiro; whether your team needs that control is the real question.

## How does each behave in Slack?

The Slack habit is nearly identical, which is why switching is cheap. In Tag, a channel member types @Claude, the response threads beneath, and the room can follow along or redirect [16]. Each thread stays bound to one session, and Claude keeps per-channel memory plus workspace notes saved from public channels [40]. Slack users do not each need a Claude seat for channel work [41]. Microsoft Teams is listed as coming soon [42].

In Jiro, a mention opens the session, the bot threads its answer, and follow-ups typed into that thread keep the session going [15]. Beyond Slack, the fleet accepts work from webhooks, schedules, the Handroll terminal, Discord and the web dashboard [43]. Jiro does not offer Tag's channel memory or channel watch; shared knowledge lives in the skillset instead [44]. Tag wins on in-Slack polish. Jiro wins on the number of other doors into the same agent.

## What about schedules, watches and webhooks?

Claude Tag's routines are scheduled jobs, channel watches and a subscription to a single pull request; there is no routine type for wider repository activity like each incoming PR [18], and schedules are capped at hourly and kept in UTC [45]. Jiro's three trigger kinds are cron for repetition, a reminder for a single later moment, and a webhook for anything external [17]. In practice: a nightly dependency check is a wash, both do it. "Tell me when anything in #incidents mentions billing" is Tag only. "Start a session when our alerting posts a webhook" is Jiro only. Teams whose work is reactive should write down which of those two sentences describes them before choosing.

## Where does the code run, and who can see it?

Tag gives each thread a short-lived sandbox in Anthropic's cloud, driven by the engine that also powers Claude Code's web product [10]. Minutes after a turn ends that sandbox is torn down; the thread, pushed branches and opened pull requests persist, but files that existed only in the sandbox are lost [46]. Anthropic keeps a transcript of each session; admins see spend per channel and an Activity page rather than individual messages, unless an Enterprise organization reads transcripts through the Compliance API [47]. Runners you host inside your own network are an option, minus admin connectors [22].

Jiro defaults to Nori's cloud; a dedicated BYOC deployment instead places private EC2 session machines under the customer's AWS control [21]. Hosted organizations are single-tenant, each with a broker of its own on a separate subdomain [48], and every session and action Nori runs is logged for search [49]. Neither vendor is weak here. Tag's advantage is Anthropic's accountability for the engine; Jiro's is that the audit trail and the machines can sit under your own account.

## How does each hand code back?

Tag's code work goes through the Claude GitHub App, which pushes branches and opens draft pull requests authored by Claude; GitLab works through a service-account token [20]. Because Claude is the author, the person who asked can approve and merge alone on a branch requiring one review, and Anthropic suggests requiring two approving reviews if a second person should look [50]. Jiro hands back a finished task, a comment, a commit or a PR, and the human reviews, merges or rejects it [19]; GitHub access comes from an installed GitHub App covering every repository or a chosen subset, or from a stored personal access token [51]. Tag supports GitLab; Jiro's documented path is GitHub. For a GitHub-only team the two are equivalent at the review step.

## When Claude Tag is the better choice

Choose Claude Tag when your organization already holds Claude Team or Enterprise seats and the finance question is settled; the seats are paid and Tag adds no second vendor [4]. Choose it when people outside engineering need to hand Claude work, because channel use needs no seat per person [41]. Choose it for setup effort close to zero: an Owner pairs the workspace once, sets a spend limit and launches, and the engine is Anthropic's own [10]. Choose it for connectors that give Claude service accounts of its own, scoped per channel, with credentials injected at the network boundary rather than placed in the sandbox [36] [37]. And choose it for routines: channel watches and standing digests that Jiro has no counterpart for [18]. Tag also supports GitLab [20]. The things to accept are the usage meter, the Team-or-Enterprise gate and the beta label [23].

## When Jiro is the better choice

Choose Jiro when your engineers are on Pro, Max or Ultra and Tag is therefore unavailable [2]; a setup token brings that plan into Jiro [33]. Choose it when finance wants one fixed number, $99 or $250 a month, with tokens on the subscription already approved [3] [5]. Choose it when anyone on the team wants Codex or Cursor in the mix, or wants to switch harness without changing vendor [7]. Choose it when repositories depend on a setup script, an MCP catalog or a shared skillset that your team maintains [11] [13]. Choose it when work should start from webhooks, Discord, a terminal or the web as well as Slack [43]. And choose it when the session machines need to run inside your own AWS account [21]. Miravoice, where better than one merged PR in four now comes from Nori, started with exactly this profile [52].

## What should a team check before deciding?

Answer these in order and stop at the first hard no. Are we on Claude Team or Enterprise? If not, Tag is unavailable [2]. Is Zero Data Retention on for our Claude organization? If so, Tag is unavailable [32]. Will finance accept a usage line that an Owner must prefund and cap [25], or do they want a flat figure [3]? Does anyone want Codex or Cursor within the year [7]? Do our repositories rely on hooks or `.mcp.json` [14]? Should webhooks from CI or alerting start work without a person typing [17]? A split result is normal. The [switching guide](https://jiro.bot/switch/claude-tag-to-jiro) describes a two-week paired pilot, and the [alternatives list](https://jiro.bot/alternatives/claude-tag) covers the wider field.

## Jiro at a glance

| Fact | Jiro |
|---|---|
| What it is | A background coding agent: it works in the cloud and returns pull requests for review [19] |
| Made by | Nori (Tilework Inc.), founded by Amol Kapoor, ex-Google Research [53] |
| Agents it runs | Claude Code, Codex, Cursor, OpenCode, pi-acp [7] |
| Whose model bill | Your Claude/ChatGPT login or API key; Cursor key [1]; Your keys, zero token markup [5] |
| Price | Free 30-day trial; Developer $99/mo; Team $250/mo [3] |
| Runtimes | Developer: up to 3; Team: 5 shared; trial: 5 for 30 days [3] [28] |
| Where work starts | Slack, web, Discord, terminal, schedules, webhooks [43] |
| Where it runs | Isolated cloud environment with your repos and tools [9]; Single-tenant per organization [48] |
| Repos | GitHub App (all or selected repos) or a PAT [51] |
| Not to be confused with | Amazon's Kiro (IDE and agent) or Atlassian's Jira (issue tracker) |
| Last verified | 2026-11-06 |

## About the publisher

Jiro is made by Nori (Tilework Inc.), a New York company. Nori was founded by Amol Kapoor, its CEO, who before Nori did AI research at Google on graph neural networks and co-authored MixHop (ICML 2019). [53] [54] Clifford Ressel is Nori's co-founder and CTO. [55] Nori's customers include Corvus and Miravoice. [56] [52]

## FAQ

### Is Jiro cheaper than Claude Tag?

It depends on volume. Tag's channel work is billed by usage from an organization balance with no published per-task price [6] [27]; Jiro adds a fixed $99 or $250 each month to a subscription you already hold [3]. Heavy usage favours the flat plan; light usage may not.

### Can Jiro use my company's Claude Team plan?

Yes. On a machine signed in to Claude, `claude setup-token` issues a token valid for a year, bound to whichever Claude plan that machine is signed into, Team included [33], and Claude Code in Jiro accepts that token [1].

### Does Claude Tag support Codex or GPT models?

No. It runs Anthropic's Opus and Sonnet models only [8]. Jiro runs Codex alongside Claude Code [7].

### Do both open pull requests?

Yes. Tag opens draft pull requests authored by Claude through the Claude GitHub App [20]. Jiro's output is a PR, commit or comment that a teammate reviews and merges [19].

### Can Claude Tag run inside our own network?

Sessions can run on runners you host, inside your own network, though connectors an admin added do not follow [22]. Jiro's answer is a dedicated BYOC deployment on AWS infrastructure you control [21].

### Is Claude Tag generally available?

Not yet. Anthropic labels it a public beta whose features may change [23]; it launched on 2026-06-23 [57].

## Try Jiro

Start the 30-day free trial, no card needed: https://login.norisessions.com/?signup=1&entrypoint=jiro

Or install Jiro's engineering skills into your own agent: `npx nori-skillsets install public/jiro`

## Sources

1. Claude Code in Nori accepts an Anthropic API key, a Claude OAuth login or a setup token; Codex accepts an OpenAI API key or Codex OAuth; Cursor Agent needs a Cursor API key; OpenCode and pi-acp take a compatible API base URL, API key and model. Source: https://docs.noriagentic.com/sessions/reference/supported-agents-and-compute (checked 2026-11-06)
2. Claude Tag is available only on Team and Enterprise plans, not on Free, Pro, Max or Ultra, and not through third-party deployments. Source: https://claude.com/docs/claude-tag/overview (checked 2026-11-06)
3. Nori's plans start with a free 30-day trial with no card; Developer is $99 a month for one user with up to three persistent runtimes, and Team is $250 a month for multiple users sharing five runtimes. Source: https://noriagentic.com/#pricing (checked 2026-11-06)
4. Anthropic's pricing page lists Claude Tag as included in the Team plan, whose standard seat costs $20 a month billed annually or $25 billed monthly, for teams of 2 to 150. Source: https://claude.com/pricing (checked 2026-11-06)
5. Nori's comparison table lists any model on your own keys with flat pricing and zero token markup. Source: https://noriagentic.com/#compare (checked 2026-11-06)
6. Claude Tag channel work is billed by usage from an organization balance rather than per seat; one-to-one DMs bill to the sender's own seat. Source: https://claude.com/docs/claude-tag/overview (checked 2026-11-06)
7. Nori's default first-class Sessions agents are Claude Code, Codex, Cursor Agent, OpenCode and pi-acp. Source: https://docs.noriagentic.com/sessions/reference/supported-agents-and-compute (checked 2026-11-06)
8. Claude Tag runs Anthropic's Opus and Sonnet models, chosen per thread, per channel or by admin default. Source: https://claude.com/docs/claude-tag/users/models (checked 2026-11-06)
9. Each Nori agent works in an isolated cloud environment with your repository, tools, dependencies and services ready to use. Source: https://noriagentic.com/#faq (checked 2026-11-06)
10. Each Claude Tag thread runs in its own ephemeral sandbox in Anthropic's cloud, on the same engine as Claude Code on the web. Source: https://claude.com/docs/claude-tag/concepts/how-it-works (checked 2026-11-06)
11. A Nori organization can define a setup script that installs dependencies and prepares every session's environment. Source: https://docs.noriagentic.com/start/what-can-nori-do#give-agents-a-real-working-environment (checked 2026-11-06)
12. A Claude Tag admin can add install commands to the setup script of the environment a channel's sessions run on, or put per-repository installs in CLAUDE.md, which repeat each session because every sandbox is fresh. Source: https://claude.com/docs/claude-tag/admins/configure-github (checked 2026-11-06)
13. Nori's dashboard exposes a catalog of MCP servers for project management, observability, analytics and other tools, and each server's own permissions still apply. Source: https://docs.noriagentic.com/sessions/guides/integrations#managed-mcp-servers (checked 2026-11-06)
14. In Claude Tag, repository hooks and .mcp.json are not loaded, effort level is not configurable, and sessions run in auto mode. Source: https://claude.com/docs/claude-tag/concepts/for-claude-code-users (checked 2026-11-06)
15. Mention Nori's bot in a Slack channel to start a session; it replies in a thread, and every reply in that thread continues the same session. Source: https://docs.noriagentic.com/sessions/guides/slack (checked 2026-11-06)
16. Claude Tag lets anyone in a Slack channel tag @Claude to hand it work; Claude replies in the thread, where the channel can see it. Source: https://claude.com/docs/claude-tag/overview (checked 2026-11-06)
17. Nori cron triggers run recurring jobs, reminders run once at a future time, and webhooks react to external events. Source: https://docs.noriagentic.com/resources/faq#can-nori-run-tasks-without-a-person-online (checked 2026-11-06)
18. Claude Tag supports scheduled routines, channel watches and single-PR subscriptions, but not routines fired by other repository events such as every new pull request. Source: https://claude.com/docs/claude-tag/users/proactivity (checked 2026-11-06)
19. Nori agents can return a pull request, commit, comment or completed task for you to review, merge or send back. Source: https://noriagentic.com/#faq (checked 2026-11-06)
20. Claude Tag code work goes through the Claude GitHub App, which pushes branches and opens draft pull requests authored by Claude; GitLab works through a service-account token. Source: https://claude.com/docs/claude-tag/admins/configure-github (checked 2026-11-06)
21. Nori hosts Jiro in its own cloud, and a dedicated BYOC deployment runs private EC2 session machines inside customer-controlled AWS infrastructure. Source: https://docs.noriagentic.com/sessions/reference/supported-agents-and-compute (checked 2026-11-06)
22. Claude Tag sessions can run on self-hosted runners inside your network, but admin-added connectors are unavailable there. Source: https://claude.com/docs/claude-tag/concepts/security-and-data (checked 2026-11-06)
23. Anthropic labels Claude Tag a public beta whose features may change before general availability. Source: https://claude.com/docs/claude-tag/admins/migrate-from-earlier (checked 2026-11-06)
24. Anthropic's Enterprise plan, which includes Claude Tag, is priced at $20 per seat a month billed annually with a 20-seat minimum, plus usage at API rates. Source: https://claude.com/pricing (checked 2026-11-06)
25. On the Team plan, Claude won't respond in channels until the organization buys usage credits. Source: https://claude.com/docs/claude-tag/admins/set-spend-limit (checked 2026-11-06)
26. When Claude Tag usage reaches the monthly spend limit, Claude stops and tells the requester in the thread that it couldn't finish; the limit counts usage at list price and resets at the start of each month. Source: https://claude.com/docs/claude-tag/admins/set-spend-limit (checked 2026-11-06)
27. Anthropic publishes no per-task cost guidance for Claude Tag and recommends running a pilot with a spend limit set, then watching the per-channel usage breakdown in admin settings. Source: https://claude.com/docs/claude-tag/admins/set-spend-limit (checked 2026-11-06)
28. Nori sizes plans by runtimes, which sleep when idle and wake on demand; the trial runs five for 30 days. Source: https://noriagentic.com/#faq (checked 2026-11-06)
29. Claude Max costs $100 a month for five times the Pro plan's usage every five hours and every week; the former Max 20x tier is now sold as Ultra. Source: https://support.claude.com/en/articles/11049741-what-is-the-max-plan (checked 2026-11-06)
30. Usage past a Pro, Max or Ultra plan's limits can continue as usage credits billed at standard API rates. Source: https://support.claude.com/en/articles/12429409-extra-usage-for-paid-claude-plans (checked 2026-11-06)
31. Claude Tag owners can set a monthly organization spend limit and per-channel limits, and Claude posts a warning in Slack as usage nears each limit. Source: https://claude.com/docs/claude-tag/admins/set-spend-limit (checked 2026-11-06)
32. Claude Tag isn't available to organizations with Zero Data Retention enabled. Source: https://claude.com/docs/claude-tag/concepts/security-and-data (checked 2026-11-06)
33. The claude setup-token command generates a one-year token tied to a Pro, Max, Ultra, Team or Enterprise plan for machines where browser login isn't available. Source: https://code.claude.com/docs/en/authentication (checked 2026-11-06)
34. Claude Tag sessions run in auto mode, where Claude's permission checker reviews each action, and an admin pre-approves routine actions with allow rules instead of a person approving in the moment. Source: https://claude.com/docs/claude-tag/concepts/for-claude-code-users (checked 2026-11-06)
35. When Claude Tag clones a granted repository, CLAUDE.md, .claude/rules/*.md and skills in .claude/skills/ load into the session, while hooks in .claude/settings.json don't run and .mcp.json is never loaded. Source: https://claude.com/docs/claude-tag/admins/configure-github (checked 2026-11-06)
36. Claude Tag admins can add connectors, plugins, skills and remote MCP servers, but local stdio MCP servers can't be connected. Source: https://claude.com/docs/claude-tag/admins/connections/custom (checked 2026-11-06)
37. Outbound traffic from an Anthropic-hosted Claude Tag sandbox is default-deny apart from GitHub, reaching only hosts covered by a credential's allowed websites, an allowed domain or the environment's network access level. Source: https://claude.com/docs/claude-tag/concepts/security-and-data (checked 2026-11-06)
38. In Nori, one claimed machine serves one active session within the organization's fleet. Source: https://docs.noriagentic.com/resources/faq#is-each-session-isolated (checked 2026-11-06)
39. A Nori skillset is written once and translated into each agent's own files: CLAUDE.md for Claude Code, GEMINI.md for Gemini CLI, copilot-instructions.md for GitHub Copilot and AGENTS.md for the rest. Source: https://docs.noriagentic.com/skillsets/concepts/translation-and-agents (checked 2026-11-06)
40. Claude Tag keeps per-channel memory and workspace notes saved from public channels, and each Slack thread stays bound to one session so replies continue it without a resume command. Source: https://claude.com/docs/claude-tag/concepts/for-claude-code-users (checked 2026-11-06)
41. Slack users don't each need a Claude seat to work with Claude Tag in channels; by default anyone in the paired Slack workspace can tag @Claude, and an Owner can restrict that to organization members. Source: https://claude.com/docs/claude-tag/overview (checked 2026-11-06)
42. Claude Tag works in Slack channels and threads for Team and Enterprise customers, and Anthropic lists Microsoft Teams as coming soon. Source: https://claude.com/product/tag (checked 2026-11-06)
43. Nori work can start from Slack, the web dashboard, Discord, the Handroll terminal, schedules and webhooks, all reaching the same fleet. Source: https://docs.noriagentic.com/start/what-can-nori-do#start-agent-work-from-the-tools-your-team-already-uses (checked 2026-11-06)
44. Nori provides the runtime around coding agents: machines, lifecycle, credentials, integrations, transcripts, checkpoints, team policy and portable skillsets. Source: https://docs.noriagentic.com/start/what-is-nori (checked 2026-11-06)
45. A Claude Tag scheduled job can run at most once an hour, Claude declines anything more frequent, and schedules run in UTC. Source: https://claude.com/docs/claude-tag/users/proactivity (checked 2026-11-06)
46. A few minutes after a Claude Tag session finishes its turn, its sandbox is released; the thread, pushed branches and opened pull requests persist, but files that existed only in the sandbox do not. Source: https://claude.com/docs/claude-tag/concepts/how-it-works (checked 2026-11-06)
47. Anthropic keeps a transcript of each Claude Tag session; admins see spend per channel and an Activity page, not individual messages, unless an Enterprise organization reads transcripts through the Compliance API. Source: https://claude.com/docs/claude-tag/concepts/how-it-works (checked 2026-11-06)
48. Each hosted Nori organization gets a single-tenant broker on its own subdomain. Source: https://docs.noriagentic.com/sessions/concepts/security-and-credentials#organization-isolation (checked 2026-11-06)
49. Nori logs every session and every action, and makes them searchable. Source: https://noriagentic.com/for-security-leaders.html (checked 2026-11-06)
50. Because Claude authors the pull requests it opens, the person who asked for a change can approve and merge it alone on a branch requiring one review; Anthropic suggests requiring two approving reviews if a second person should look. Source: https://claude.com/docs/claude-tag/admins/configure-github (checked 2026-11-06)
51. Nori supports two GitHub credential methods: a GitHub App you install and scope to all repositories or a selected set, or a saved personal access token. Source: https://docs.noriagentic.com/sessions/guides/integrations#github (checked 2026-11-06)
52. Nori handles more than 25% of Miravoice's merged pull requests, and Miravoice's CTO says it easily saves him five hours a week. Source: https://noriagentic.com/newsletter/2026-09-18-miravoice-case-study.html (checked 2026-11-06)
53. Nori is led by CEO Amol Kapoor, a former Google Research scientist, and CTO and co-founder Clifford Ressel. Source: https://ai.engineer/speakers/amol-kapoor (checked 2026-11-06)
54. Amol Kapoor's Google-affiliated research includes MixHop (ICML 2019), N-GCN and a 2020 spatio-temporal graph neural network for COVID-19 forecasting. Source: https://scholar.google.com/citations?user=SC_-qB8AAAAJ (checked 2026-11-06)
55. Clifford Ressel is Nori's CTO and co-founder. Source: https://ai.engineer/orgs/nori (checked 2026-11-06)
56. Corvus's CTO saw a production bug at a client dinner, had Nori look into it and deploy a fix. Source: https://noriagentic.com/newsletter/2026-05-19-corvus-case-study.html (checked 2026-11-06)
57. Anthropic announced Claude Tag on 2026-06-23 as a beta for Claude Enterprise and Team customers. Source: https://www.anthropic.com/news/introducing-claude-tag (checked 2026-11-06)

---

Jiro is a background coding agent by Nori. It is not Amazon's Kiro or Atlassian's Jira.
