Kontext CLI: Credential Broker for AI Coding Agents

What Kontext CLI Does
Kontext CLI is an open-source command-line tool that wraps AI coding agents to provide credential management without exposing API keys. It addresses the problem of teams copy-pasting long-lived API keys into .env files or chat interfaces, which creates secret sprawl and lacks access lineage.
How It Works
You declare what credentials a project needs in a .env.kontext file with placeholders like:
GITHUB_TOKEN={{kontext:github}}
STRIPE_KEY={{kontext:stripe}}
LINEAR_TOKEN={{kontext:linear}}Then run kontext start --agent claude. The CLI authenticates you via OIDC and exchanges placeholders for credentials:
- For services supporting OAuth: short-lived access tokens via RFC 8693 token exchange
- For static API keys: credentials injected directly into the agent's runtime environment
Secrets exist only in memory during the session — never written to disk on your machine. The backend holds OAuth refresh tokens and API keys; the CLI never sees them, only getting back short-lived access tokens scoped to the session.
Key Features
- One command to launch Claude Code:
kontext start --agent claude - Ephemeral credentials: short-lived tokens scoped to the session, automatically expired on exit
- Declarative credential templates in
.env.kontextfiles - Governance telemetry: Claude hook events streamed to backend with user, session, and org attribution
- Secure by default: OIDC authentication, system keyring storage, RFC 8693 token exchange
- Lean runtime: native Go binary (~5ms hook overhead per tool call), uses ConnectRPC for backend communication
- Update notifications on
kontext start(cached for 24h, disable withKONTEXT_NO_UPDATE_CHECK=1)
Installation and Usage
Install with: brew install kontext-dev/tap/kontext
Or direct binary install:
tmpdir="$(mktemp -d)" \ && gh release download --repo kontext-dev/kontext-cli --pattern 'kontext_*_darwin_arm64.tar.gz' --dir "$tmpdir" \ && archive="$(find "$tmpdir" -maxdepth 1 -name 'kontext_*_darwin_arm64.tar.gz' -print -quit)" \ && tar -xzf "$archive" -C "$tmpdir" \ && sudo install -m 0755 "$tmpdir/kontext" /usr/local/bin/kontext
From any project directory with Claude Code installed: kontext start --agent claude
On first run, the CLI handles everything interactively — login, provider connections, credential resolution. Clear stored OIDC session with kontext logout.
Audit and Governance
The CLI captures for every tool call: what the agent tried to do, what happened, whether it was allowed, and who did it — attributed to a user, session, and org. Every tool call is streamed for audit as the agent runs.
Works with Claude Code today, Codex support coming soon. Server-side policy enforcement is in development — the infrastructure for allow/deny decisions on every tool call is already wired.
📖 Read the full source: HN AI Agents
👀 See Also

Strale.io offers free IBAN and email validation API for AI agents with no signup
Strale.io provides a free API with five capabilities including IBAN validation, email validation, DNS lookup, URL-to-markdown conversion, and JSON repair. No signup or API key is required, and it includes an MCP server for Claude or Cursor integration.

Claude Code Ultracode Mode Spawns 70-Agent Pipeline for Deep Search
A single 'deep search' request in Claude Code's ultracode mode auto-generated a 4-phase pipeline with ~70 agents, each fetching and cross-checking projects independently. The orchestrator script keeps intermediate results out of the context window, preventing context overload.

The Human Creativity Benchmark: Separating Convergence from Divergence in AI Creative Evaluation
Contra Labs introduces the Human Creativity Benchmark (HCB), a framework that distinguishes objectively verifiable criteria (e.g., prompt adherence) from subjective taste (e.g., visual appeal) in evaluating generative AI for creative work. The benchmark reveals that no current model is reliably both correct and steerable, addressing mode collapse and the need for differentiated output.

OpenClaw Integrates Features from Claude Code Leak
An OpenClaw user had their bot analyze the leaked Claude Code (Rust recreation by Instructkr) and selectively ported specific architectural patterns into their OpenClaw setup. The integration focuses on practical improvements like automatic startup continuity, conversation compaction, and a pre-tool/post-tool hook framework.