tinker-ai-skill (fine-tune on your sessions)
A skill that walks your agent through turning your own Claude Code (or Codex/OpenCode) session logs into a fine-tuning dataset, training a LoRA on Tinker, merging it and exporting a GGUF you can run locally in llama.cpp. The authors say it is the pipeline behind their Tess-4 open models. MIT.
Test results
Run with our skill harness: the skill alone in a throwaway project, invoked by name, and the real output file checked. Skills are tested in Claude Code and Cursor only so far.
| Client | Result | Last passed | What we checked |
|---|---|---|---|
| Claude Code | Not tested yet | never | |
| Cursor 2026.11.17-f532c45 | Passed Tested 2026-11-17 | 2026-11-17 | tinker-ai-skill: {'anonymized/*.jsonl written by the pipeline': True, 'task kept (units request still in the data)': True, 'build ref kept (not over-redacted)': True, 'no name/email/handle/org/username left': True, 'fake |
Skill version: github.com/drost-ai/tinker-ai-skill @ c8d199b8 (2026-11-07).
Tested on ToolPassed badge
Maker of tinker-ai-skill (fine-tune on your sessions)? It passed in Cursor on 2026-11-17. Add this free badge to your README; it links back to this listing. Free, no strings: the badge never changes a result.
[](https://toolpassed.com/skills/tinker-ai-skill/)Get it
The skill’s files and install instructions are in its repository: github.com/drost-ai/tinker-ai-skill.
Not working?
Fixes from our own test runs: Skill not working in Cursor?