autofl-nvflare-report
Skill1k starsChanged 14 days agoApache-2.0View on GitHub
Generate and commit a markdown report after an Auto-FL NVFlare autoresearch experiment has been manually stopped. Use when the user asks to summarize a stopped campaign, report achieved improvements, explain implemented literature-derived ideas and sources, refresh progress plots, capture pasted agent model/effort/cost context when available, or commit the final report and progress plot to the current experiment branch.
Trending
Less talk lately.about the whole repository
Who's talking about NVIDIA/NVFlare
- Kaggle/kaggle-sdk-python“kagglesdk 0.1.31 and 0.1.32 import a missing competitions.legacy package”GitHub · Jun 30, 2026 · 1 reactions
- londonaicentre/FLIP“docs: call the NVIDIA framework FLARE, its short official name”GitHub · Oct 17, 2026
- NVIDIA/SkillEvaluator“[BUG]: Generated BENCHMARK.md does not faithfully explain a passing gate and splits canonical case IDs”GitHub · Oct 9, 2026
- Denimworld12/Hospital-Chain“feat: run hospital nodes on separate machines with provisioning kits and a cross-address verifier”GitHub · Oct 2, 2026
- ai4os/ai4os-nvflare“Update NVFlare version to 2.8”GitHub · Jul 16, 2026
The biggest 5 of 25 mentions linking to the repository that our crawl found. Each line is their own headline.
See all 25
Talked about on GitHub (25 times).
On GitHub · 25
- Jun 30, 2026kagglesdk 0.1.31 and 0.1.32 import a missing competitions.legacy package — Kaggle/kaggle-sdk-python1 reaction · 2 comments
- Oct 17, 2026docs: call the NVIDIA framework FLARE, its short official name — londonaicentre/FLIP0 reactions · 1 comment
- Oct 9, 2026[BUG]: Generated BENCHMARK.md does not faithfully explain a passing gate and splits canonical case IDs — NVIDIA/SkillEvaluatorsays it tested it0 reactions · 1 comment
- Oct 2, 2026feat: run hospital nodes on separate machines with provisioning kits and a cross-address verifier — Denimworld12/Hospital-Chain0 reactions
- Jul 16, 2026Update NVFlare version to 2.8 — ai4os/ai4os-nvflare0 reactions
20 more on GitHub
- Jul 6, 2026Upgrade NVFlare backbone 2.7.2 → 2.8.0 (deferred until a clean 2.7.2 run) — KatherLab/MediSwarm0 reactions · 3 comments
- Jun 26, 2026Upgrade NVFLARE backend 2.7.2 → 2.8.0 — londonaicentre/FLIP0 reactions
- Jun 22, 2026Add step-ca ephemeral admin certs — pcnudde/NVFlare0 reactions · 1 comment
- May 7, 2026feat(privacy): federated learning architecture-ready (Horizon Europe / EU4Health alignment) — Yanstart/VarunaPoC0 reactions · 1 comment
- Jan 7, 2026T3: Model fitting (deploying GWAS tool / model on every client) [Enamul, ++] — collaborativebioinformatics/FedGen0 reactions
- Sep 10, 2025[Feature]: From Vertical FL feasibility (#363) to first example — londonaicentre/FLIP0 reactions · 1 comment
- Jun 23, 2025Secure xgboost example — RhinoHealth/user-resources0 reactions · 3 comments
- May 14, 2025NVFlare backbone upgrade for better connection performance — KatherLab/MediSwarm0 reactions · 2 comments
- Mar 24, 2025NVFlare Integration — KCL-BMEIS/UltraFlwr0 reactions
- Jul 14, 2024Try to fix experiment already exist issue in `MLFlowHandler` — Project-MONAI/MONAI0 reactions · 11 comments
- Sep 13, 2023Add stats_sender to MonaiAlgo for FL stats — Project-MONAI/MONAI0 reactions · 1 comment
- Aug 9, 2023【CS-part3】New submissions for Wed, 9 Aug 23 — Yukeaaa/arxiv-daily0 reactions
- Jul 12, 2023New learner cifar10 — yanchengnv/NVFlare0 reactions
- May 6, 2023Bump nvflare from 2.0.10 to 2.3.0 in /federated_learning — NVIDIA/fsi-samples0 reactions · 1 comment
- Mar 22, 2023[WIP] add `FLStatsHandler` — Project-MONAI/MONAI0 reactions · 3 comments
- Mar 21, 2023Try NVFlare cifar10 example — KatherLab/swarm-learning-hpe0 reactions
- Mar 21, 2023Apply the new data split strategies — KatherLab/swarm-learning-hpe0 reactions
- Dec 6, 2022How to use bst.eval_set() and bst.update() with xgboost_ray — ray-project/xgboost_ray0 reactions · 3 comments
- May 18, 2022713 Update nvflare example for 2.1 — Project-MONAI/tutorials0 reactions · 2 comments
- Mar 22, 2022Bump nvflare from 2.0.2 to 2.0.16 in /federated_learning/nvflare/nvflare_spleen_example — Project-MONAI/tutorials0 reactions · 4 comments
In this repository16 other files it gives its agents.
- AGENTS.mdAGENTS.md
- examples/tutorials/self-paced-training/AGENTS.mdAGENTS.md
- research/auto-fl-research/AGENTS.mdAGENTS.md
- CLAUDE.mdCLAUDE.md
- research/auto-fl-research/CLAUDE.mdCLAUDE.md
- autofl-nvflareSkill
- nvflare-autofl-reportSkill
- nvflare-autoflSkill
- nvflare-convert-huggingfaceSkill
- nvflare-convert-lightningSkill
- nvflare-convert-pytorchSkill
- nvflare-diagnose-jobSkill
- nvflare-fed-statsSkill
- nvflare-orientSkill
- nvflare-sharedSkill
- nvflare-example-skillSkill
The file
We split this skill into the rules it gives an agent
10 rules. About 0 are common, 4 are distinctive, and 6 are very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈4
Distinctive · estimate
This author's own choices. Worth a look.
6
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Distinctive: line 8
Use this skill after the human has manually stopped an autoresearch campaign.
- Very specific: line 10
Do not launch new experiments.
- Distinctive: line 16
If the branch is main, do not auto-commit unless the user explicitly asked to commit to main.
- Very specific: line 18
Use the human-provided PYTHON interpreter when one is specified.
- Distinctive: line 27
If plotting fails because of local font/cache paths, retry with writable cache env vars, for example:
- Very specific: line 39
Keep this separate from experiment runtime cost.
- Very specific: line 51
If no context files exist, use --agent-settings "<summary>" and --agent-cost "<summary>" instead.
- Distinctive: line 63
Commit the report and progress plot automatically to the current experiment branch:
- Very specific: line 70
If a non-default plot path was used, stage that plot path instead of progress.png.
- Very specific: line 71
Do not stage run_logs/, NVFlare result directories, generated local caches, or unrelated edits.
≈ “Common” is an estimate: at least 1 in 5 of the 372,429 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
Auto-FL NVFlare Report
Use this skill after the human has manually stopped an autoresearch campaign.
Do not launch new experiments. The job is to summarize the completed branch from results.tsv, existing artifacts, source notes, and any available pasted agent model/effort/cost context, then commit the report markdown and refreshed progress plot.
Workflow
- Verify context:
- Confirm the current branch with
git status -sb. - If the branch is
main, do not auto-commit unless the user explicitly asked to commit tomain. - Confirm
results.tsvexists and has rows. - Use the human-provided
PYTHONinterpreter when one is specified; do not create virtual environments or install dependencies unless the user explicitly asks. - Preserve unrelated dirty files; stage only the generated report and refreshed progress plot unless the user asks otherwise.
- Confirm the current branch with
Refresh the progress plot:
"${PYTHON:-python3}" scripts/plot_progress.py results.tsv --output progress.pngIf plotting fails because of local font/cache paths, retry with writable cache env vars, for example:
MPLCONFIGDIR=/tmp/mpl-cache XDG_CACHE_HOME=/tmp/xdg-cache \ "${PYTHON:-python3}" scripts/plot_progress.py results.tsv --output progress.png- Capture agent model/effort/cost context when available:
- Claude Code:
/model,/effort, and/costare interactive slash commands and cannot be invoked from a shell or tool call. If the human pasted those outputs into the prompt or saved them to text files, pass the exact text via--agent-settings,--agent-settings-file,--agent-cost, or--agent-cost-file. - Claude Code without pasted model/effort output: pass
--agent-settings "Agent model/effort telemetry unavailable in this Claude Code runtime; /model and /effort are interactive and were not provided to the reporting agent." - Claude Code without pasted cost output: pass
--agent-cost "Agent cost telemetry unavailable in this Claude Code runtime; /cost is interactive and was not provided to the reporting agent. Experiment runtime cost is reported from results.tsv." - Codex: if no runtime model/effort/cost command is exposed, pass explicit unavailable notes with
--agent-settingsand--agent-cost. - Keep this separate from experiment runtime cost. Runtime cost comes from
results.tsvand measures aggregate candidate execution time, not agent token spend.
- Claude Code:
Generate the report:
"${PYTHON:-python3}" skills/autofl-nvflare-report/scripts/generate_report.py \ --results results.tsv \ --plot progress.png \ --agent-settings-file /tmp/autofl-agent-settings.txt \ --agent-cost-file /tmp/autofl-agent-cost.txtIf no context files exist, use
--agent-settings "<summary>"and--agent-cost "<summary>"instead. The script prints the report path. By default it writes toreports/<branch>-autoresearch-report.md.- Review the generated report for obvious parsing issues:
- best score and baseline are present;
- progress plot is embedded as a markdown image and points to the refreshed plot path;
- final recommended stack includes exact budget/args;
- runtime section includes total aggregate runtime and average runtime per candidate;
- agent/tooling context section is present and either contains pasted model/effort/cost output or explains why it was unavailable;
- literature-derived ideas include source refs from
[src: ...]markers when present; - the report explicitly distinguishes candidates from reproduced/kept results.
Commit the report and progress plot automatically to the current experiment branch:
git add <report-path> progress.png git commit -m "Add autoresearch campaign report"If a non-default plot path was used, stage that plot path instead of
progress.png. Do not stagerun_logs/, NVFlare result directories, generated local caches, or unrelated edits.
Report Expectations
The report should lead with an executive summary, then include a technical appendix. It must cover:
- baseline score, best score, absolute and relative lift;
- embedded progress plot image;
- best run index, description, commit, budget, runtime, status, and artifact path;
- major running-best milestones and likely improvement mechanisms;
- final recommended stack and whether it preserves the current contract or uses an explicit protocol mode such as SCAFFOLD;
- total aggregate candidate runtime, average runtime per candidate, crash count, and failure notes;
- agent model/effort settings and agent/tooling cost if provided by the human, or explicit unavailable notes;
- literature-derived ideas, source refs, and whether each helped;
- null/worse/unstable ideas that should not be retried;
- recommended reproduction and next experiment directions.
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
No reviews yet
People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.
- It worked
- Partly
- It didn't
Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

