# TABStack Codex API Evaluation

**Category:** Search
**Score:** 69 (C)

This is the agent-readable Devtool Arena evaluation page for TABStack on Codex API.

## Summary

| Metric | Value |
|--------|-------|
| Status | completed |
| Overall score | 69 |
| Grade | C |
| Eval score | 62 |
| Discovery score | 86 |
| Cost | $0.79 |
| Runtime | 6m 28s |
| Tool calls | 31 |
| Errors | 10 |
| Tokens used | 5607883 |
| Success | Yes |

## Agent-Readiness Checklist

| Signal | Evidence |
|--------|----------|
| Context7 | Not found |
| llms.txt | https://tabstack.ai/llms.txt |
| MCP server | https://tabstack.ai/integrations/claude-agent |
| Typed SDK | https://pypi.org/project/tabstack/ |
| OpenAPI | https://tabstack.ai/openapi.json |
| Agent skills | https://tabstack.ai/integrations/claude-agent |
| CLI | https://github.com/mozilla/pilo |

## Evaluation Prompt

````text

# Task
Using https://docs.tabstack.ai/api, complete the following task:

Using TABStack, build me a script that runs a small web research workflow:

1. Run TWO related searches:
   a. A broad query: "grid-scale battery storage advances"
   b. A more specific, recency-focused query: "grid-scale battery storage breakthroughs 2025"
2. Merge the results from both searches, deduplicate by URL, and select the top 5 most relevant results
3. If TABStack supports fetching/extracting page content, retrieve the full content of the single most relevant result and write a 2-3 sentence summary grounded in that fetched content (not just the snippet)
4. Print the output as JSON with two keys:
   - "results": the top 5 deduplicated results, each with title, url, and snippet (or description)
   - "summary": your grounded 2-3 sentence summary of the top result (include the source url) 

## Execution
After creating the script, run it to verify it works:
```bash
cd /home/daytona/app && python <your_script>.py
```

The script should print output to stdout. If there are errors, debug and fix them until it runs successfully.

````

## Grader Results

| Check | Passed | Weight | Score | Details |
|-------|--------|--------|-------|---------|
| Cost | No | 0 | 0 | $0.79 ($0.70–1.00, very expensive) |
| Time | No | 0 | 0 | 388s (5–10min, very slow) |
| Efficiency | No | 0 | 0 | 31 tool calls (>25, inefficient) |
| Found Docs | Yes | 0 | — | Touched docs at docs.tabstack.ai: yes (16 matching tool calls) |
| Zero Errors | No | 0 | — | Tool outputs with errors/tracebacks: 10 |
| Syntax Valid | Yes | 0 | — | Code compiles/parses correctly |
| Created Files | Yes | 0 | — | Generated files: 2 (need ≥1) |
| Full Execution | Yes | 0 | — | Method: full_execution, Exit: 0 |

## Run Artifacts

| Artifact | Value |
|----------|-------|
| Conversation turns | 38 |
| Tool call traces | 31 |
| Generated files | 2 |
| Exit code | 0 |
| Completed at | 2026-09-29T16:57:51.801087+00:00 |

### Generated Files

- /home/daytona/app/requirements.txt
- /home/daytona/app/battery_research.py

## Related Pages

- [Codex API leaderboard](/leaderboard/codex/api)
- [Compare Codex API companies](/leaderboard/codex/api/compare)
- [Agent Landscape](/leaderboard/discoverability)

Canonical URL: https://devtoolarena.com/codex/api/tabstack