Do coding agents recommend Instabase?
Coding agents raised Instabase in 60 of 801 judged document processing sessions and chose it in none of them.
Coding agents raised Instabase in 60 of 801 judged document processing sessions and chose it in none of them.
This page reports what happened when Claude Code, Codex, Cursor, Grok Build CLI and Muse Code had to solve a problem in document processing inside a realistic codebase. Not what a chat assistant says about Instabase. What an agent actually installed.
The numbers
| Category | Document processing |
| Sessions in the category | 801 |
| Sessions where Instabase was chosen | 0 |
| Install share | 0% |
| Raised as a candidate, not chosen | 60 |
| Chosen when considered | 0% |
| Site | instabase.com |
Which agents raised Instabase
Claude Code, Codex, Cursor and Grok Build CLI considered it and none adopted it. Grok Build CLI raised it most often, in 13% of its document processing sessions.
| Agent | Sessions | Raised Instabase | Chose it |
|---|---|---|---|
| Claude Code | 180 | 20 (11%) | 0 |
| Codex | 235 | 14 (6%) | 0 |
| Cursor | 180 | 22 (12%) | 0 |
| Grok Build CLI | 30 | 4 (13%) | 0 |
| Muse Code | 176 | 0 (0%) | 0 |
Which buyers it came up for
It surfaced most for requests written as enterprise team, in 24% of those sessions. Knowing which buyer already has the product in mind tells you which pages to fix first.
| Who is asking | Sessions | Raised it |
|---|---|---|
| Vibe coder | 68 | 0 (0%) |
| Junior developer | 327 | 16 (5%) |
| Senior engineer | 270 | 12 (4%) |
| Enterprise team | 136 | 32 (24%) |
What Instabase was up against
The full ranking in document processing, from the same sessions:
| # | Product | Runs won | Share |
|---|---|---|---|
| 1 | Azure AI Document Intelligence | 193 | 24% |
| 2 | Anthropic Claude | 150 | 19% |
| 3 | Amazon Textract | 103 | 13% |
| 4 | Google Cloud Document AI | 49 | 6% |
| 5 | OpenAI models | 46 | 6% |
| 6 | Google Gemini | 45 | 6% |
| 7 | Docling | 24 | 3% |
| 8 | Apache PDFBox | 23 | 3% |
What this means
Raised 60 times and chosen none is a specific, diagnosable result, and a better starting position than being unknown. The agent has the product in mind. It reached the shortlist 60 times. Something then lost every one of those sessions.
In our data the causes, in the order they occur:
- The quickstart does not run when pasted, so the agent abandoned it mid-integration.
- The documentation describes an interface that changed, so the generated code failed.
- The product name and the package name differ, so the install step went wrong.
- The fit was genuinely wrong for the repository, which is fine and worth knowing.
All but the last are fixable in days, and the reason is written down in the session transcript.
Where these numbers come from
The 801 sessions in document processing are part of a published set of 15,000, run with real coding agents inside realistic codebases and judged blind. The full method is on one page: how we measured this.
Every document processing run can be replayed on the board.
If you work on Instabase: the judge recorded a reason for every session where it was raised and passed over. Those reasons are in the transcripts.
<!-- generated by scripts/write-data-pages.mjs -->
Common questions
Do coding agents recommend Instabase?
They raise it but do not choose it. Across 60 sessions where Instabase came up as a candidate, agents chose something else every time.
Does Claude Code recommend Instabase?
In 0 of the 180 sessions in document processing run with Claude Code, which is 0%.
Do different coding agents treat Instabase differently?
Not much. Claude Code, Codex, Cursor, Grok Build CLI and Muse Code chose it at similar rates, between 0% and 0% of their runs.
How was this measured?
Real coding agents at pinned versions were run in sandboxes inside 92 realistic codebases and asked to solve real tasks. A simulated project owner approved or questioned each recommendation before any code was written, and a judge from a model family that builds none of the agents read every session blind.
Where this comes from
Armature ran 15,000 judged sessions with Claude Code, Codex, Cursor, Grok Build CLI and Muse Code inside 92 realistic codebases, and published every run. The numbers on this page come from that work.