Kiro: what 38 developers actually say.

38 Hacker News opinions from 30 developers, Nov 2025 – Nov 2026. 26 come from people who say they use it. Here is what they praise, what they complain about, and who switched.

38 opinions30 developersNov 2025 – Nov 2026Methodology
Approval
59%
±13.1 at 90% confidence
Last 90 days
–
No recent data
Rank
#14
of 22 tools · CI-adjusted
Opinions
38
26 first-hand
Developers
30
1.3 opinions each
The short version

The case for Kiro, in one paragraph.

Kiro earns 59% approval across 38 opinions. Developers are divided on it.

Pick it if you

  • Want a spec-first workflow that can run an hour or more on a task
  • Like the Kiro CLI and want remote or Telegram access to an agent
  • Want strong spec generation and predictable cost per dollar

Look elsewhere if you

  • Need Cursor-level IDE features and polish
  • Work in large monorepos where harness reliability matters
  • Worry about context resets mid-task or unreviewed AI-written code

Where it wins and loses.

Mentions across 38 opinions. Bars are counts, not votes; a tool can be both praised and criticized on the same topic.

TopicComplaints ← → PraiseMentions
UX 11 mentions
38
PraiseComplaint
1 topics shown · sorted by net sentiment · source: 38 opinions
View data table
TopicPraiseComplaintsMentions
UX8311

How opinion has moved.

Monthly approval, Dec 2025 – Jun 2026. Bars show opinion volume; low-volume months are still included but faded.

33% Dec 202533% Dec 2025 low79% Jun 2026– last 90 days
0255075100APPROVAL59% all-time33%79%Dec '25Feb '26Apr '26Jun '26
Monthly approval, 0–100. Bars show opinion volume. Months with fewer than 3 opinions are left out.
View data table
MonthApprovalOpinions
Dec 202533%3
Jan 202660%5
Feb 202683%3
Mar 202660%5
Apr 202643%7
May 202658%6
Jun 202679%7

What developers say.

Summaries are paraphrased. Each links to the original Hacker News comment.

Praise

4 selected
UX

They switched to Kiro, finding it nearly identical but with better visibility into what the harness does.

HN comment · Jul 7, 2026Read on HN ↗
Reliability

Finds Kiro surprisingly good; would switch fully if saving and resuming were more reliable.

HN comment · Jun 22, 2026Read on HN ↗
UX

Kiro is fine and they prefer it to other tools they have tried.

HN comment · Jun 22, 2026Read on HN ↗
UX

They like Kiro CLI and find it easier to follow than Claude Code.

HN comment · Jun 9, 2026Read on HN ↗

Complaints

4 selected

Tried Kiro and found most of the Cursor features they like missing.

HN comment · Aug 31, 2026Read on HN ↗
UX

The Kiro IDE is poor, though the Kiro CLI was acceptable for a CLI.

HN comment · May 5, 2026Read on HN ↗

Kiro is not in the same league as Cursor.

HN comment · Apr 22, 2026Read on HN ↗
UX

Uses Kiro at Amazon, whose harness is the main gripe; says Claude Code is much better.

HN comment · Mar 15, 2026Read on HN ↗

Questions developers ask.

Is Kiro worth it?
59% of 38 opinions are positive, so reactions are mixed. Fans point to spec generation, harness visibility, and the $20 plan. Critics point to missing features and failed upgrades on large repos. Try it on a real task before committing.
Is Kiro better than Claude Code?
The data leans no. 3 people said they prefer Claude Code over Kiro, and 3 named Claude Code as the tool they moved to. One Amazon user called Claude Code much better. A few others do prefer Kiro for CLI readability.
Is Kiro better than Cursor?
Opinions differ. One user said Kiro is not in the same league as Cursor, and another missed most Cursor features. Others said they would switch fully if saving and resuming worked better. IDE polish is the main gap.

The data behind this page.

Developers are split on Kiro. Approval sits at 59% across 38 opinions from 30 authors, with 26 sharing firsthand experience. Praise centers on visibility into what the harness does, the CLI experience, and clear value on the $20 plan, while complaints cite missing Cursor-level features, reliability problems, and failed large refactors. Monthly approval is noisy on small samples: 43% in April, 79% in June, then 0% on a single opinion in August. Claude Code is the main alternative, named by 3 people as something they prefer and 3 as something they switched from.

Every number comes from public Hacker News comments collected through the HN Search API. A language model tagged each comment for sentiment, first-hand use, topic and switches; a human spot-checks samples. No vendor can pay for placement.

20,064 comments screened38 about Kiro26 first-hand

Read the full methodology →