Agent skill

Sft Job Cleanup

by open-thoughts in open-thoughts/OpenThoughts-Agent

Publish + clean up a finished LLaMA-Factory SFT job on a no-internet HPC cluster (Jupiter/Leonardo): cancel pending retries, drop intermediate checkpoints, HF-upload the model to its configured…

Apache-2.0Auto-check passedAI & LLM Engineering

Install Sft Job Cleanup

skills CLI

$ npx skills add open-thoughts/OpenThoughts-Agent --skill sft-job-cleanup -a claude-code

Project install by default; add -g for ~/.claude/skills/.

GitHub CLI

$ gh skill install open-thoughts/OpenThoughts-Agent sft-job-cleanup --agent claude-code

Project scope by default; add --scope user for a personal install. Needs GitHub CLI 2.90.0 or later (public preview).

Manual copy

$ git clone --depth 1 https://github.com/open-thoughts/OpenThoughts-Agent.git skills-src && mkdir -p .claude/skills && cp -r skills-src/.agents/skills/sft-job-cleanup .claude/skills/sft-job-cleanup && rm -rf skills-src

Use ~/.claude/skills/ instead of .claude/skills for a personal install. The folder must contain SKILL.md.

Claude Code skills documentation · loads skills from .claude/skills/

Paste into Claude Code

Install the "sft-job-cleanup" agent skill from https://github.com/open-thoughts/OpenThoughts-Agent/tree/main/.agents/skills/sft-job-cleanup into .claude/skills/sft-job-cleanup/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "sft-job-cleanup", then confirm the skill loads.

Claude Code copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.

$skill-installer

$skill-installer install https://github.com/open-thoughts/OpenThoughts-Agent/tree/main/.agents/skills/sft-job-cleanup

Type this inside Codex. $skill-installer <name> installs a curated skill from openai/skills. The installer writes to $CODEX_HOME/skills (default ~/.codex/skills). Restart Codex if the skill does not show up.

skills CLI

$ npx skills add open-thoughts/OpenThoughts-Agent --skill sft-job-cleanup -a codex

Project install goes to .agents/skills/; add -g for ~/.codex/skills/.

GitHub CLI

$ gh skill install open-thoughts/OpenThoughts-Agent sft-job-cleanup --agent codex

Project scope by default (.agents/skills/); add --scope user for a personal install.

Manual copy

$ git clone --depth 1 https://github.com/open-thoughts/OpenThoughts-Agent.git skills-src && mkdir -p .agents/skills && cp -r skills-src/.agents/skills/sft-job-cleanup .agents/skills/sft-job-cleanup && rm -rf skills-src

Use ~/.agents/skills/ instead of .agents/skills for a personal install.

Codex skills documentation · loads skills from .agents/skills/

Paste into Codex

Install the "sft-job-cleanup" agent skill from https://github.com/open-thoughts/OpenThoughts-Agent/tree/main/.agents/skills/sft-job-cleanup into .agents/skills/sft-job-cleanup/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "sft-job-cleanup", then confirm the skill loads.

Codex copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.

skills CLI

$ npx skills add open-thoughts/OpenThoughts-Agent --skill sft-job-cleanup -a cursor

Project install goes to .agents/skills/; add -g for ~/.cursor/skills/.

GitHub CLI

$ gh skill install open-thoughts/OpenThoughts-Agent sft-job-cleanup --agent cursor

Project scope by default (.agents/skills/); add --scope user for a personal install.

Manual copy

$ git clone --depth 1 https://github.com/open-thoughts/OpenThoughts-Agent.git skills-src && mkdir -p .cursor/skills && cp -r skills-src/.agents/skills/sft-job-cleanup .cursor/skills/sft-job-cleanup && rm -rf skills-src

Use ~/.cursor/skills/ instead of .cursor/skills for a personal install.

Cursor skills documentation · loads skills from .cursor/skills/, .agents/skills/, .claude/skills/, .codex/skills/

Paste into Cursor

Install the "sft-job-cleanup" agent skill from https://github.com/open-thoughts/OpenThoughts-Agent/tree/main/.agents/skills/sft-job-cleanup into .cursor/skills/sft-job-cleanup/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "sft-job-cleanup", then confirm the skill loads.

Cursor copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.

gemini skills install

$ gemini skills install https://github.com/open-thoughts/OpenThoughts-Agent.git --path .agents/skills/sft-job-cleanup

--scope user (default) or --scope workspace; --path is the subfolder of the repo that holds the skill; --consent skips the security confirmation prompt.

skills CLI

$ npx skills add open-thoughts/OpenThoughts-Agent --skill sft-job-cleanup -a gemini-cli

Project install goes to .agents/skills/; add -g for ~/.gemini/skills/.

GitHub CLI

$ gh skill install open-thoughts/OpenThoughts-Agent sft-job-cleanup --agent gemini-cli

Project scope by default (.agents/skills/); add --scope user for a personal install.

Manual copy

$ git clone --depth 1 https://github.com/open-thoughts/OpenThoughts-Agent.git skills-src && mkdir -p .gemini/skills && cp -r skills-src/.agents/skills/sft-job-cleanup .gemini/skills/sft-job-cleanup && rm -rf skills-src

Use ~/.gemini/skills/ instead of .gemini/skills for a personal install, then run /skills reload.

Gemini CLI skills documentation · loads skills from .gemini/skills/, .agents/skills/

Paste into Gemini CLI

Install the "sft-job-cleanup" agent skill from https://github.com/open-thoughts/OpenThoughts-Agent/tree/main/.agents/skills/sft-job-cleanup into .gemini/skills/sft-job-cleanup/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "sft-job-cleanup", then confirm the skill loads.

Gemini CLI copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.

GitHub CLI

$ gh skill install open-thoughts/OpenThoughts-Agent sft-job-cleanup

Installs for Copilot at project scope by default; add --scope user for a personal install. Preview a skill first with gh skill preview. Needs GitHub CLI 2.90.0 or later (public preview).

skills CLI

$ npx skills add open-thoughts/OpenThoughts-Agent --skill sft-job-cleanup -a github-copilot

Project install goes to .agents/skills/; add -g for ~/.copilot/skills/.

Manual copy

$ git clone --depth 1 https://github.com/open-thoughts/OpenThoughts-Agent.git skills-src && mkdir -p .github/skills && cp -r skills-src/.agents/skills/sft-job-cleanup .github/skills/sft-job-cleanup && rm -rf skills-src

Use ~/.copilot/skills/ instead of .github/skills for a personal install. Commit .github/skills so cloud agent and code review can use it.

GitHub Copilot skills documentation · loads skills from .github/skills/, .claude/skills/, .agents/skills/

Paste into GitHub Copilot

Install the "sft-job-cleanup" agent skill from https://github.com/open-thoughts/OpenThoughts-Agent/tree/main/.agents/skills/sft-job-cleanup into .github/skills/sft-job-cleanup/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "sft-job-cleanup", then confirm the skill loads.

GitHub Copilot copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.

skills CLI

$ npx skills add open-thoughts/OpenThoughts-Agent --skill sft-job-cleanup -a opencode

OpenCode documents no install command of its own. Project install goes to .agents/skills/; add -g for ~/.config/opencode/skills/.

GitHub CLI

$ gh skill install open-thoughts/OpenThoughts-Agent sft-job-cleanup --agent opencode

Project scope by default (.agents/skills/); add --scope user for a personal install.

Manual copy

$ git clone --depth 1 https://github.com/open-thoughts/OpenThoughts-Agent.git skills-src && mkdir -p .opencode/skills && cp -r skills-src/.agents/skills/sft-job-cleanup .opencode/skills/sft-job-cleanup && rm -rf skills-src

Use ~/.config/opencode/skills/ instead of .opencode/skills for a personal install.

OpenCode skills documentation · loads skills from .opencode/skills/, .claude/skills/, .agents/skills/

Paste into OpenCode

Install the "sft-job-cleanup" agent skill from https://github.com/open-thoughts/OpenThoughts-Agent/tree/main/.agents/skills/sft-job-cleanup into .opencode/skills/sft-job-cleanup/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "sft-job-cleanup", then confirm the skill loads.

OpenCode copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.

Facts

Skill name: sft-job-cleanup
GitHub stars: 301
Token cost: ~1.9k tokens
SKILL.md length: 591 words
Files: 1
Skills in repo: 44
Repo updated: Sep 28, 2026
First seen: Oct 7, 2026
Licence: Apache-2.0

At a glance

Publish + clean up a finished LLaMA-Factory SFT job on a no-internet HPC cluster (Jupiter/Leonardo): cancel pending retries, drop intermediate checkpoints, HF-upload the model to its configured…

An SFT fine-tune finishes and needs uploading + registering
SKILL.md covers Recognition heuristic — which…, Cross-cutting upload rules…, 8B SFT Job Cleanup Checklist and 32B SFT Job Cleanup Checklist…, plus 1 more section
Calls hf and python
Run the SFT cleanup checklist

What it does

Sft Job Cleanup is an agent skill from open-thoughts/OpenThoughts-Agent. Publish + clean up a finished LLaMA-Factory SFT job on a no-internet HPC cluster (Jupiter/Leonardo): cancel pending retries, drop intermediate checkpoints, HF-upload the model to its configured --hubmodelid, register in Supabase via manualdbpush (--training-type SFT default), and free disk. Covers the 8B path (root safetensors, direct upload), the 32B/ZeRO-3 path (consolidate shards → safetensors first), the Qwen3.5 preprocessorconfig copy, the don't-upload-partials policy, and the hf-upload gotchas (tmux not…

Its SKILL.md is about 1.9k tokens, which your agent loads only when the skill is triggered. It is a single SKILL.md file with no bundled scripts.

It sits in AI & LLM Engineering. It works with Qwen, tmux and Supabase. The repository describes itself as: Data recipes and robust infrastructure for training AI agents. The licence is Apache-2.0.

When your agent uses it

An SFT fine-tune finishes and needs uploading + registering
Run the SFT cleanup checklist

Example prompts

“run the SFT cleanup checklist”
“/sft-job-cleanup”

Requirements

Python 3

What it can do on your machine

Read from SKILL.md and the folder at commit 3bd1917. It shows what the files ask for, not the result of running them.

Tool permissions
Pre-approves nothing: there is no allowed-tools line, so your agent's usual permission prompts apply.
From allowed-tools in the SKILL.md frontmatter.
Runs code
Shell commands in SKILL.md call:
- hf
- python
From the folder's file list and the shell code blocks in SKILL.md.
Network
No URLs in SKILL.md.
From URLs in SKILL.md, links to its own repository left out.
Credentials
Names no API keys, tokens, secrets or passwords.
From names ending in _API_KEY, _TOKEN, _SECRET, _KEY or _PASSWORD in SKILL.md.

Context cost

Sft Job Cleanup loads about 1.9k tokens when it runs. Until then it costs about 204 tokens; SKILL.md has 591 words of instructions outside code blocks.

Always · name and description, kept in context so the agent knows when to use it: ~204
When it runs · the whole SKILL.md, loaded when a task matches: ~1.9k

Estimates: characters ÷ 4, the usual rule of thumb; real counts depend on the model's tokenizer. Scripts and assets cost tokens only if the agent reads them.

Safety

Auto-check passed

The automated check found no risky patterns in SKILL.md.

Automated static check — not a guarantee. Review scripts before installing. It scans the text of SKILL.md for risky patterns (piping downloads into a shell, reading credential files, hidden Unicode, destructive commands); files beside SKILL.md are not scanned.

SKILL.md

The full file from open-thoughts/OpenThoughts-Agent at commit 3bd1917, republished under its Apache-2.0 licence (© open-thoughts). 591 words, ~1,858 tokens.

Download SKILL.mdSave it as .claude/skills/sft-job-cleanup/SKILL.md (or your agent's skills folder).

name: sft-job-cleanup
description: Publish + clean up a finished LLaMA-Factory SFT job on a no-internet HPC cluster (Jupiter/Leonardo): cancel pending retries, drop intermediate checkpoints, HF-upload the model to its configured --hub_model_id, register in Supabase via manual_db_push (--training-type SFT default), and free disk. Covers the 8B path (root safetensors, direct upload), the 32B/ZeRO-3 path (consolidate shards → safetensors first), the Qwen3.5 preprocessor_config copy, the don't-upload-partials policy, and the hf-upload gotchas (tmux not nohup, `hf upload` not `-large-folder`, Leonardo sbatch-tunnel not login node). Use when an SFT fine-tune finishes and needs uploading + registering, or "run the SFT cleanup checklist". Distinct from RL cleanup (rl-agentic-job-cleanup) and datagen cleanup (datagen-job-cleanup).

sft-job-cleanup

After an SFT job completes on Jupiter or Leonardo, publish the model and clean up.

Recognition heuristic — which path? (check the checkpoint root first)

bash

ls $CHECKPOINTS_DIR/<job_name>/ | grep -E 'safetensors|global_step'

Root model-*.safetensors → 8B path (including Qwen3.5; no consolidation).
global_stepN/ plus zero_to_fp32.py, without root safetensors → 32B path (consolidate ZeRO-3 shards).

Cross-cutting upload rules (apply to both paths)

hf upload, NEVER hf upload-large-folder (deprecated stub + deadlocks on HF LFS 429s). Wrap any non-trivial upload in tmux, not nohup/disown.
--private is a no-value flag — omit it (default public); --private false is a CLI parse error.
Jupiter: login node has direct internet → hf upload from the login node (in tmux) works.
Leonardo: the login node SIGKILLs long processes at ~100s → use the sbatch compute-node + SSH-tunnel upload (.agents/ops/leonardo/ops.md "Leonardo HF Upload — Use sbatch, NOT the Login Node" / sft-launch).
Don't upload partials: if training is below 100%, relaunch and auto-resume. Salvage-upload only with explicit approval, as laion/<job_name>-<step>-<size>.

Tokenizer sanity check: tokenizer_config.json extra_special_tokens must be a dict, not a list; replace a list with {} before upload.

bash

python -c "import json;d=json.load(open('<ckpt>/tokenizer_config.json'));assert isinstance(d.get('extra_special_tokens',{}),dict), 'LIST — coerce to {}'"

8B SFT Job Cleanup Checklist

0. Cancel pending retries (so stale restarts don't fire mid-upload):

bash

squeue -u $USER --format='%i %j %T' | grep <job_name> | grep PENDING | awk '{print $1}' | xargs -r scancel

1. Remove intermediate checkpoints (don't upload cruft):

bash

rm -rf $CHECKPOINTS_DIR/<job_name>/checkpoint-*  $CHECKPOINTS_DIR/<job_name>/.cache

1b. Qwen3.5 only — copy preprocessor_config.json from the base model:

bash

cp /path/to/Qwen3.5-9B/preprocessor_config.json  $CHECKPOINTS_DIR/<job_name>/   # or the -27B base

2. Upload model weights to HuggingFace. Naming: full final upload (training reached 100%) → the configured --hub_model_id from the launch command (laion/<descriptive_name>, NO step/size suffix — do NOT use the job name verbatim). (Partial salvage, only-if-OK'd → laion/<job_name>-<step>-<size>.)

bash

# Jupiter login node (direct internet). On LEONARDO use the §11 sbatch-tunnel — login-node hf upload dies at ~100s.
source ~/secrets.env
tmux new-session -d -s hf_upload_<short> \
    "source ~/secrets.env && hf upload <hub_model_id> $CHECKPOINTS_DIR/<job_name> . \
        --repo-type=model 2>&1 | tee $CHECKPOINTS_DIR/<job_name>/upload.log"
# tmux attach -t hf_upload_<short>  (Ctrl-b d to detach)

Wait for it to finish and verify the repo exists on HF Hub.

3. Register in the unified DB (SFT is the DEFAULT --training-type, no flag needed):

bash

python scripts/database/manual_db_push.py \
  --hf-model-id <hub_model_id> --base-model <base_model_hf> \
  --dataset-name <dataset_name>          # comma-separated for multi-dataset → sets dataset_names

SKIP for HF-only series (e.g. Delphi #6279 — YAMLs set enable_db_registration: false; do not register, and do not pass an anchor as --base-model since that auto-creates a base-model row).

4. Clean up the experiments dir — only after 1–3 succeed:

bash

rm -rf $EXPERIMENTS_DIR/<job_name>

Show full SKILL.md (296 more words)Show less

32B SFT Job Cleanup Checklist (DeepSpeed ZeRO-3 — consolidate first)

For 32B ZeRO-3 SFT without stage3_gather_16bit_weights_on_model_save: true, consolidate shards before upload.

0. Cancel pending retries (same as 8B).

1. Verify training reached 100% — trainer_log.jsonl shows current_steps == total_steps. Default policy: don't salvage partials (relaunch + resume); only proceed if explicitly OK'd as a partial.

2. Consolidate ZeRO-3 shards → fp32 state_dict → safetensors:

bash

python -m hpc.launch --job_type consolidate \
  --consolidate_input $CHECKPOINTS_DIR/<job_name> \
  --consolidate_output_repo <hub_model_id> \
  --consolidate_workdir <writable_workdir>/<job_name> \
  --time_limit 02:00:00 --num_nodes 1

Produces <workdir>/<job_name>/final_repo/ with root-level weights, tokenizer, and config. Do not rely on its final HF push; manually upload after final_repo/ is complete.

3. Manually upload from final_repo/ (NOT the original checkpoint dir — it still holds ZeRO-3 shards). Naming same as 8B (full → --consolidate_output_repo/--hub_model_id, no suffix):

bash

# Jupiter login node. On LEONARDO use the §11 sbatch-tunnel (131GB → ~4 min). tmux; hf upload (not -large-folder).
source ~/secrets.env
tmux new-session -d -s hf_upload_<short> \
    "source ~/secrets.env && hf upload <hub_model_id> <consolidate_workdir>/<job_name>/final_repo . \
        --repo-type=model 2>&1 | tee <consolidate_workdir>/<job_name>/upload.log"

4. Register in the unified DB (same as 8B step 3; SFT is the default; skip for HF-only series).

5. Clean up — only after 2–4 succeed, remove BOTH the sharded checkpoint dir AND the consolidate workdir (32B sharded ckpt ~700GB + workdir ~200GB):

bash

rm -rf $CHECKPOINTS_DIR/<job_name>  <consolidate_workdir>/<job_name>

Launch-side details (preamble, configs, sbatch patching, the no-internet pre-download) live in the sft-launch skill (per-cluster particulars in ops/<cluster>/ops.md §SFT); this skill is the post-run publish + cleanup.

Operating notes

Run full cleanup after a completed, 100% job: cancel pending chain, drop checkpoints, upload, register, and clean the experiment directory. Flag obvious anomalies; cancellation of running jobs remains user-driven.
Multi-dataset DB registration: pass the full comma-separated list to --dataset-name so dataset_names is populated (not just one dataset_id). Known limitation: the script stores it as a single string and does NOT trigger the multiple_datasets path (dataset_id ends up null) — verify the right field after registering. Single-dataset --dataset-name works fine and populates dataset_id.
Baseline model versioning (Sera/CoderForge): flat monotonic -v5/-v6/-v7 in HF repo names + README iteration tables, NOT nested v4-v2/v4-v3. In-flight runs keep their existing names; the NEXT retrain uses the new scheme (next Sera = v5, skipping v4 to avoid colliding with existing v4 artifacts).

Files

Just SKILL.md in .agents/skills/sft-job-cleanup of open-thoughts/OpenThoughts-Agent.

Open the folder on GitHubat commit 3bd1917

Compare with similar skills

Sft Job Cleanup next to the 5 skills that share the most tags, products or categories with it. Stars are the repository's; “used in” counts other GitHub owners with a copy.

Sft Job Cleanup compared with similar skills
Skill	Stars	Used in	Tokens	Auto-check	Licence	Repo updated
Sft Job Cleanup this skillopen-thoughts/OpenThoughts-Agent	301	—	~1.9k	Automated safety check: Pass	Apache-2.0	12 days ago
Agent Feature ReproductionQwenLM/qwen-code	28k	—	~1.5k	Automated safety check: Pass	Apache-2.0	today
tmux Real User TestingQwenLM/qwen-code	28k	—	~2.3k	Automated safety check: Pass	Apache-2.0	today
Fix Art IssuesOpenPipe/ART	11k	—	~840	Automated safety check: Notes	Apache-2.0	today
Diffusion Perf Optvllm-project/vllm-omni	7.1k	—	~7.5k	Automated safety check: Pass	Apache-2.0	today
Qwen Mtp GgufR6410418/Jackrong-llm-finetuning-guide	1.7k	—	~1.7k	Automated safety check: Pass	MIT	3 mo ago

Similar skills

Agent Feature Reproduction
QwenLM/qwen-code
Reproduces a feature from Codex or Claude Code in Qwen Code by running the reference agent under capture, reading the traces, then implementing matching behavior.
28k GitHub stars~1.5k tokensUpdated today
DevelopmentAuto-check passed
tmux Real User Testing
QwenLM/qwen-code
Drives Qwen Code in a real tmux session the way a user would and saves a readable step-by-step transcript of each screen for maintainers to review.
28k GitHub stars~2.3k tokensUpdated today
Testing & QAAuto-check passed
Fix Art Issues
OpenPipe/ART
Fix a GitHub issue on OpenPipe/ART and open a PR. An agent skill from OpenPipe/ART.
11k GitHub stars~840 tokensUpdated today
AI & LLM EngineeringAuto-check: notes
Diffusion Perf Opt
vllm-project/vllm-omni
Diagnose and optimize vLLM Omni diffusion workloads, especially Wan/Qwen/Flux-style image and video generation.
7.1k GitHub stars~7.5k tokensUpdated today
AI & LLM EngineeringAuto-check passed
Qwen Mtp Gguf
R6410418/Jackrong-llm-finetuning-guide
Complete agent-ready workflow for Qwen-family MTP or nextn GGUF conversion and release.
1.7k GitHub stars~1.7k tokensUpdated 3 mo ago
AI & LLM EngineeringAuto-check passed
Dy Note
Rimagination/dy-note
DyNote: systematically and efficiently extract raw Douyin/DY video data and analyze videos, comments, accounts, hashtags, and short-video scenes into evidence-graded learning notes, summaries…
172 GitHub stars~4.5k tokensUpdated 3 mo ago
AI & LLM EngineeringAuto-check passed

More from open-thoughts/OpenThoughts-Agent

All 44 skills in this repo

Analyze Dataset Token Length
open-thoughts/OpenThoughts-Agent
Analyze the token length of an OT-Agent conversation-format (ShareGPT-style) dataset — the per-trace distribution (median/p90/max) and/or counts under a token threshold + a metadata predicate (e.g.
301 GitHub stars~1.5k tokensUpdated 12 days ago
Auto-check passed
Analyze Id Eval Ranking
open-thoughts/OpenThoughts-Agent
Given a list of models (HF name stubs) that have valid agentic ID eval scores in Supabase, build a ranking table: raw per-benchmark accuracy on the 3 ID benchmarks (SWE-Bench-100…
301 GitHub stars~3.1k tokensUpdated 12 days ago
Auto-check passed
Analyze Job History Iris
open-thoughts/OpenThoughts-Agent
Run the Iris harbor job-history analyzer (scripts/iris/analyzeirisharborjob.py) on a datagen/eval job and read its JSON sidecar for trustworthy throughput / preemption / productive-trial stats.
301 GitHub stars~2.9k tokensUpdated 12 days ago
Auto-check passed
Analyze Rl Behavior
open-thoughts/OpenThoughts-Agent
Run the full RL behavioral-analysis pipeline (scripts/analysis/analyzerlbehavior.py) on a trained RL model to understand WHAT changed vs its pre-RL baseline, WHY, whether it PERSISTS, and its EVAL…
301 GitHub stars~4.2k tokensUpdated 12 days ago
Auto-check passed
Analyze Training Run Iris
open-thoughts/OpenThoughts-Agent
Detailed health check for a Levanter/executor TRAINING run on the marin Iris cluster (e.g.
301 GitHub stars~2k tokensUpdated 12 days ago
Auto-check passed
Code Create Staged Plan
open-thoughts/OpenThoughts-Agent
DESIGN a non-trivial codebase change (Harbor / MarinSkyRL / vLLM / OT-Agent / LLaMA-Factory) as a dependency-ordered STAGED PLAN before writing code — a feature port, a multi-step fix with parity…
301 GitHub stars~1.5k tokensUpdated 12 days ago
Auto-check passed

Works with

Questions about Sft Job Cleanup

What does Sft Job Cleanup do?

Publish + clean up a finished LLaMA-Factory SFT job on a no-internet HPC cluster (Jupiter/Leonardo): cancel pending retries, drop intermediate checkpoints, HF-upload the model to its configured…. Sft Job Cleanup is an agent skill from open-thoughts/OpenThoughts-Agent. Publish + clean up a finished LLaMA-Factory SFT job on a no-internet HPC cluster (Jupiter/Leonardo): cancel pending retries, drop intermediate checkpoints, HF-upload the model to its configured --hubmodelid, register in Supabase via manualdbpush (--training-type SFT default), and free disk.

When should I use Sft Job Cleanup?

Sft Job Cleanup fits situations like: an SFT fine-tune finishes and needs uploading + registering; run the SFT cleanup checklist.

How do I install Sft Job Cleanup in Claude Code?

Run `npx skills add open-thoughts/OpenThoughts-Agent --skill sft-job-cleanup -a claude-code`. Or copy the skill folder (.agents/skills/sft-job-cleanup in open-thoughts/OpenThoughts-Agent) into .claude/skills/sft-job-cleanup in your project. Claude Code loads it when a task matches its description.

How do I install Sft Job Cleanup in Codex?

Run `npx skills add open-thoughts/OpenThoughts-Agent --skill sft-job-cleanup -a codex`. Or copy the skill folder (.agents/skills/sft-job-cleanup in open-thoughts/OpenThoughts-Agent) into .agents/skills/sft-job-cleanup in your project. Codex loads it when a task matches its description.

Can I use Sft Job Cleanup in Cursor, Gemini CLI or GitHub Copilot?

Cursor, Gemini CLI, GitHub Copilot and OpenCode also load SKILL.md folders. With the skills CLI, run `npx skills add open-thoughts/OpenThoughts-Agent --skill sft-job-cleanup -a cursor` (or -a gemini-cli, github-copilot or opencode for the others). To copy it by hand, put the folder in .cursor/skills/sft-job-cleanup, .gemini/skills/sft-job-cleanup, .github/skills/sft-job-cleanup and .opencode/skills/sft-job-cleanup in your project.

What does Sft Job Cleanup need to run?

Going by SKILL.md and its folder, Sft Job Cleanup needs the command-line tools its instructions call (hf and python). Our summary lists: Python 3.

Does Sft Job Cleanup access the network?

SKILL.md contains no URLs. Any network use would come from the scripts or tools the agent runs. This is read from the text; nothing was executed.

Is Sft Job Cleanup safe to install?

Our automated static check of SKILL.md found no risky patterns, such as piping downloads into a shell, reading credential files or hidden Unicode. It is not a guarantee. Review the folder before installing.

What licence does Sft Job Cleanup use?

Sft Job Cleanup is published under the Apache-2.0 licence (the repository's licence). It allows redistribution, so the full SKILL.md is shown on this page.

How many tokens does Sft Job Cleanup use?

About 1.9k tokens (SKILL.md is roughly 7.4k characters). Agents keep only the skill's name and description in context until a task matches; then they load SKILL.md in full.

What are the alternatives to Sft Job Cleanup?

Skills that share tags, products or a category with Sft Job Cleanup: Agent Feature Reproduction (QwenLM/qwen-code, 28k stars), tmux Real User Testing (QwenLM/qwen-code, 28k stars), Fix Art Issues (OpenPipe/ART, 11k stars) and Diffusion Perf Opt (vllm-project/vllm-omni, 7.1k stars). The comparison table on this page puts their stars, adoption, token cost, safety result and licence side by side.

Who maintains Sft Job Cleanup?

open-thoughts (a GitHub organization) maintains it in open-thoughts/OpenThoughts-Agent, which has 301 GitHub stars. The repository holds 44 skills in this directory. The repository was last updated on September 28, 2026.

Source: open-thoughts/OpenThoughts-Agent on GitHub. Facts on this page come from the repository at the commit we read; the author's words are quoted as theirs.

Install Sft Job Cleanup

Facts

At a glance

What it does

When your agent uses it

Example prompts

Requirements

What it can do on your machine

Tool permissions

Runs code

Network

Credentials

Context cost

Safety

SKILL.md

sft-job-cleanup

Recognition heuristic — which path? (check the checkpoint root first)

Cross-cutting upload rules (apply to both paths)

8B SFT Job Cleanup Checklist

32B SFT Job Cleanup Checklist (DeepSpeed ZeRO-3 — consolidate first)

Operating notes

Files

Compare with similar skills

Similar skills

More from open-thoughts/OpenThoughts-Agent

Related

Questions about Sft Job Cleanup