eksec

Executive Summary

Rebrand notice (June 2026): eksec is now Nairi. As of 2026-06-11, eksec.ai returns an HTTP 308 permanent redirect to nairi.ai,^[1] and the project's open-source daemon retains backwards compatibility with legacy EKSEC_API_KEY configuration and ~/.config/eksecd/ paths.^[2] This profile retains the eksec name for continuity but reflects the Nairi product.

Nairi (formerly eksec.ai) is a team agent deployment platform — "Your team's personal agent."^[1] The product takes underlying coding agents — Claude Code, OpenCode, or Codex — and deploys them as accessible team bots in Slack, Discord, a web app, or via API. This enables non-technical team members to interact with sophisticated AI agents using plain English in the chat tools they already use. Since this profile was first published in February 2026, the company has rebranded, published transparent pricing, added GitHub repo integration, and open-sourced its agent-runner daemon under MIT.^[2]

---

Product Overview

Nairi positions itself as the deployment layer for team-accessible AI agents. Rather than building another coding agent, it focuses on making existing agents (Claude Code, OpenCode, Codex) accessible to entire teams through familiar chat interfaces.^[1]

Key Capabilities

Product Surfaces / Editions

---

Technical Architecture

Nairi operates as an orchestration layer between chat platforms and coding agents:

Slack/Discord/Web/API → Nairi → Claude Code / OpenCode / Codex
                          ↓
                  nairid daemon (hosted or self-hosted)
                          ↓
                  MCP Servers (databases, tools), GitHub repos

Setup Flow

1. Choose your base — Select Claude Code, OpenCode, or Codex as the underlying agent harness
2. Set it up — Configure MCPs, skills, and rules to control agent capabilities
3. Share it — Connect to Slack, Discord, the web app, or use the API^[1]

Key Technical Details

---

Use Cases

Nairi highlights several primary use cases on its website:^[1]

Data Analyst

Connect a read-only database or data warehouse. Team members ask questions in plain English and get insights without SQL knowledge — no tickets, no waiting.

AI SRE / On-Call Incident Response

Wire up logs, database, and codebase. When production issues occur, team members can investigate in real-time through chat. The agent traces root causes and can even submit fix PRs.

Code Reviewer

Trigger the agent from CI on every pull request. It reviews diffs, leaves inline GitHub comments, and catches issues before human reviewers — without adding another SaaS dashboard.

Second Brain / Customer Support

Newer use cases added since launch: a "company brain" on Slack that answers questions from internal knowledge, and customer support agents.^[1]

---

Strengths

• Non-technical accessibility — Anyone on the team can interact with coding agents via Slack/Discord, democratizing AI assistance beyond developers
• Agent flexibility — Not locked to one provider; choose between Claude Code, OpenCode, or Codex based on task needs
• MCP-native — 1,000+ MCP tools enable database queries, API calls, and custom tool access^[1]
• Transparent pricing with no per-seat costs — Flat plans from $20/month; the Team plan allows unlimited teammates^[4]
• Open-source daemon — The MIT-licensed nairid runner allows self-hosted agents and shows rapid release cadence (114 releases, latest 2026-06-10)^[2]
• Customer validation — Named testimonials from Yespark, Wecasa, Misfits, Talkpush, and Depozen indicate real production usage^[1]

---

Cautions

• Fresh rebrand — The eksec → Nairi rename (eksec.ai now 308-redirects to nairi.ai) creates discoverability churn; existing references, docs, and integrations point at the old name^[1]
• Early-stage viability risk — Funding status and team size remain unknown; the daemon's modest traction (23 GitHub stars as of June 2026) reflects a small community^[2]
• Security questions — Deploying agents with database access to entire teams raises access control concerns; the site mentions isolated containers per agent, but no SOC 2 or detailed security documentation is public^[1]
• No independent reviews — All public quotes are first-party testimonials; no significant Hacker News, Reddit, or third-party coverage found as of June 2026

---

What Developers Say

No independent community discussion (Hacker News, Reddit, or third-party reviews) of eksec or Nairi was found as of June 2026. The following are first-party testimonials published on the Nairi website, quoted verbatim:^[1]

"Nairi has revolutionized how our company works...the level of adoption has been amazing." — Guillaume Wrobel, Co-founder, Yespark^[1]

"Nairi drastically improved our day-to-day workload...speed of execution has been a huge value add." — Hugo Vast, SRE, Wecasa^[1]

"The most magical one has been Nairi giving our Ops team natural language access...significantly reduced." — Chaitanya Dhawan, CBO & Co-founder, Misfits^[1]

"It was exactly what I needed." — Romain Verbeke, Head of Engineering, Talkpush^[1]

Treat these as vendor-curated; independent validation is not yet available.

---

Pricing & Licensing

Pricing became public after this profile was first published. As of June 2026:^[4]

All plans include a 14-day free trial. The self-hosted nairid daemon is MIT-licensed.^[2]

---

Competitive Positioning

Direct Competitors

When to Choose Nairi Over Alternatives

• Choose Nairi when: You want non-technical team members to access coding agents via Slack/Discord, with flat no-per-seat pricing
• Choose Tembo when: You need developer workflow integrations (GitHub, Linear, Sentry) and multi-repo operations
• Choose Runbear when: You want no-code agent building with simpler LLM backends
• Choose Devin when: You want a fully autonomous AI engineer, not agent deployment
• Choose OpenWork when: You need a fully open-source, self-hosted platform

---

Ideal Customer Profile

Best fit:

• Teams where non-developers need to interact with AI agents (support, sales, ops)
• Companies wanting ad-hoc database querying without SQL training
• Organizations with Slack/Discord as primary communication
• Teams comfortable with early-stage products

Poor fit:

• Enterprise buyers needing SOC 2, SSO, and detailed security documentation
• Organizations wary of products mid-rebrand
• Teams wanting a fully self-hosted platform (only the agent daemon is self-hostable)

---

Viability Assessment

Nairi is shipping fast — the rebrand, public pricing, GitHub integration, and a near-daily daemon release cadence all landed within roughly four months of this profile's first publication. But funding, team, and security posture remain opaque, and there is no independent validation of the product yet.

---

Bottom Line

Nairi (formerly eksec.ai) solves a specific problem: making coding agents accessible to non-technical team members via Slack and Discord. By wrapping Claude Code, OpenCode, or Codex in a chat-friendly interface, it enables ad-hoc database queries, production debugging, and code reviews without requiring users to understand the underlying AI systems. The June 2026 picture is healthier than February's: transparent pricing ($20–$80/month, no per-seat costs), an MIT-licensed self-hosted daemon, GitHub repo support, and five named customer testimonials.

Recommended for: Teams where non-developers need AI assistance — sales querying databases, support investigating issues, ops debugging production — and Slack/Discord is the primary workspace, especially given the flat unlimited-teammates pricing.

Not recommended for: Enterprises needing SOC 2, SSO, and mature vendor stability, or organizations that require independent third-party validation before adopting tooling.

Outlook: The rebrand to Nairi plus public pricing and an open-source daemon signal a company maturing past stealth, and release velocity is genuinely high. The niche between no-code agent builders (Runbear) and developer orchestration platforms (Tembo) remains interesting. Watch for funding disclosure, security certifications, and whether independent community adoption follows the first-party testimonials.

---

Research by Ry Walker Research • methodology