Kora — Universal API CLI
1. Project Summary
| Field | Value |
|---|---|
PRJ ID |
PRJ-2026-04-001 |
Date Proposed |
2026-04-01 |
Owner |
Evan Rosado |
Priority |
P0 (Critical) |
Category |
Revenue |
Status |
Proposed |
Target Completion |
Phase 1: June 2026 / Phase 2: September 2026 |
Estimated Effort |
300-400 hours (Phase 1-3) |
Repository |
|
2. Problem Statement
Every engineer who works with APIs faces three universal problems:
-
Authentication is painful. Every API has different auth — OAuth2 flows, API keys in different headers, bearer tokens that expire, mTLS, session cookies. Engineers waste hours configuring auth for each vendor, each tool, each project.
-
There is no man page for APIs. When you SSH into a box, you type
--helpand learn the tool. With APIs, you read 200 pages of browser docs, then copy-paste curl commands. There’s no interactive discovery from the terminal. -
API output is ugly and unusable. Every API returns raw JSON. Engineers pipe through
jq, write one-off scripts, copy to spreadsheets. There’s no standard way to get a clean table, CSV, or filtered view.
These problems affect every developer, network engineer, security analyst, SRE, DevOps, and cloud engineer who touches an API. That’s millions of engineers daily.
3. Proposed Solution
Kora — a compiled Go CLI that provides unified, vendor-aware, interactive API access. One tool to authenticate, discover, query, and format responses from any API. Named after the West African 21-stringed instrument: one instrument, many strings, one musician makes them all sing.
4. Value Proposition
4.1. Who Benefits
| User | Pain Point Solved |
|---|---|
Network Engineer |
|
Security Analyst |
|
DevOps Engineer |
|
Cloud Engineer |
|
SRE |
|
Developer |
|
4.2. Why Now
-
Go + Cobra ecosystem is mature (kubectl, docker, gh, terraform all use it)
-
Charm TUI libraries make beautiful terminals possible (Lipgloss, Bubbletea)
-
API-first architecture is standard — every product exposes REST endpoints
-
No existing tool unifies vendor-aware API access with credential management
-
Personal: 23 vendors already implemented in netapi (Python) — architecture proven, ready for Go port
-
Personal: RTX 5090 hardware + Ollama stack can power AI-assisted API discovery (future differentiator)
4.3. Success Metrics
| Metric | Target | Measurement |
|---|---|---|
GitHub Stars (6 months) |
500-2,000 |
GitHub API |
Unique users (12 months) |
1,000+ |
Homebrew install counts, binary downloads |
Vendor implementations |
10 at launch, 30 by month 12 |
Community + core |
Monthly sponsorship revenue (12 months) |
$1,000-3,000 |
GitHub Sponsors + Open Collective |
First enterprise customer (18 months) |
1 |
Signed contract |
6. Improvement Proposals
|
Proposals from ecosystem audit — 2026-04-04. For team review and prioritization. |
| Priority | Proposal | Rationale | Effort |
|---|---|---|---|
P1 |
Add quickstart guide with concrete examples |
13 partials cover architecture and business case but zero hands-on usage. A developer cannot start using kora without reading the source code. |
M |
P1 |
Add field-notes partial (technical decisions, lessons) |
Commercial ambitions require documented rationale for technology choices, API design decisions, and pivot points. |
M |
P2 |
API endpoint reference table |
Universal API CLI needs a structured reference: endpoint, method, auth type, payload, response schema per vendor. |
L |
P2 |
Authentication flow documentation |
Multi-vendor auth (API keys, OAuth, certificates) is the hardest part. Document the flow per vendor with sequence diagrams. |
M |
P3 |
Competitive analysis update |
Market positioning section exists but needs refresh against current CLI landscape (httpie, restish, curlie, etc.). |
S |
7. Metadata
| Field | Value |
|---|---|
PRJ ID |
PRJ-2026-04-001 |
Author |
Evan Rosado |
Date Created |
2026-04-01 |
Last Updated |
2026-04-01 |
Status |
Proposed |
Next Review |
2026-04-15 |