Individual skill
ai-research-reproduction
Rigor Reproduce compatible skill slug for README-first deep learning repository reproduction. Use when the user wants an end-to-end, minimal-trustworthy flow that reads the repository first, selects the smallest documented inference or evaluation target, coordinates intake, setup, trusted execution, optional trusted training, optional repository analysis, and optional paper-gap resolution, enforces conservative patch rules, records evidence assumptions deviations and human decision points, and writes the standardized `repro_outputs/` bundle. Do not use for paper summary, generic environment setup, isolated repo scanning, standalone command execution, silent protocol changes, score chasing, or broad research assistance outside repository-grounded reproduction.
skills/ai-research-reproduction/SKILL.md
Source commit fb3ccdf5aa64b1fa5b1d0463e666a8bc8bc26e01
What was checked
- Matches the observed repository HEAD
- Yes
- Scanner rules are current
- No
- Repository scan complete
- No
- Skill scan complete
- No
- Observed behavior
- Not checked yet
- Last checked
- Nov 5, 2026
These are observations at the recorded commit, not a guarantee of safety. Review the source before running it.
Install this skill
Choose an agent in your Stack or copy a command below. Commands fetch the pinned commit and copy only this skill directory. Existing destinations are refused. Copying a command does not install anything.
An install command is unavailable: current rules, complete repository and skill scans, an eligible behavior level and a permitted license are required.
Repository context
- Repository stars (not skill ratings)
- 486
- Repository last updated
- Sep 23, 2026
- License
- MIT
Observed agent integration: No specific integration evidence. Standard SKILL.md installation does not imply verified compatibility.
- agent-skills Repository topic
- ai-agents Repository topic
- deep-learning Repository topic
- reproducibility Repository topic
- research-tools Repository topic
- research Derived category
- docs Derived category
- education Derived category
- testing Derived category
- testing-browser Derived category
Files and excerpt
File metadata and permitted, redacted excerpts come from the recorded scan. Source links are pinned to that commit.
Browse files
SKILL.mdcore · 7,123 bytesSKILL.mdcore · 7,123 bytesSKILL.mdcore · 7,123 bytes_bundled/MANIFEST.jsonasset · 5,394 bytes_bundled/shared/scripts/agent_provider.pyruns · 6,289 bytes_bundled/shared/scripts/command_utils.pyruns · 3,085 bytes_bundled/shared/scripts/lessons_store.pyruns · 8,608 bytes_bundled/shared/scripts/model_adapter.pyruns · 6,487 bytes_bundled/shared/scripts/resource_monitor.pyruns · 8,164 bytes_bundled/shared/scripts/runtime_runner.pyruns · 29,154 bytes_bundled/shared/scripts/task_queue.pyruns · 30,927 bytes_bundled/shared/scripts/write_explore_bundle.pyruns · 33,409 bytes_bundled/shared/scripts/write_run_bundle.pyruns · 31,741 bytes_bundled/skills/analyze-project/scripts/analyze_project.pyruns · 25,957 bytes_bundled/skills/env-and-assets-bootstrap/scripts/plan_setup.pyruns · 4,795 bytes_bundled/skills/env-and-assets-bootstrap/scripts/prepare_assets.pyruns · 4,288 bytes_bundled/skills/minimal-run-and-audit/scripts/write_outputs.pyruns · 1,171 bytes_bundled/skills/repo-intake-and-plan/scripts/extract_commands.pyruns · 13,256 bytes_bundled/skills/repo-intake-and-plan/scripts/scan_repo.pyruns · 2,937 bytes_bundled/skills/run-train/scripts/run_training.pyruns · 19,876 bytes_bundled/skills/run-train/scripts/write_outputs.pyruns · 1,173 bytesagents/openai.yamlasset · 389 bytesassets/COMMANDS.template.mddoc · 348 bytesassets/LOG.template.mddoc · 669 bytesassets/PATCHES.template.mddoc · 525 bytesassets/SUMMARY.template.mddoc · 621 bytesassets/model_profile.template.jsonasset · 183 bytesassets/readme_annotations.template.jsonasset · 1,101 bytesassets/status.template.jsonasset · 1,170 bytesreferences/agent-job.mddoc · 5,670 bytesreferences/agent-operating-principles.mddoc · 2,768 bytesreferences/agent-runner.mddoc · 10,444 bytesreferences/architecture.mddoc · 1,322 bytesreferences/continuous-learning-policy.mddoc · 3,825 bytesreferences/deep-learning-experiment-principles.mddoc · 2,077 bytesreferences/explore-variant-spec.mddoc · 4,215 bytesreferences/language-policy.mddoc · 794 bytesreferences/output-spec.mddoc · 9,248 bytesreferences/patch-policy.mddoc · 1,678 bytesreferences/readme-annotation-policy.mddoc · 4,403 bytesreferences/research-pitfall-checklist.mddoc · 1,665 bytesreferences/research-rigor-principles.mddoc · 3,754 bytesreferences/research-safety-principles.mddoc · 1,925 bytesreferences/research-thinking-loop.mddoc · 4,189 bytesreferences/runtime-and-model-adapter.mddoc · 5,470 bytesscripts/annotate_readme.pyruns · 36,767 bytesscripts/doctor.pyruns · 3,621 bytesscripts/orchestrate_repro.pyruns · 118,106 bytesscripts/repro_job.pyruns · 19,806 bytesscripts/run_agent.pyruns · 39,314 bytes
Read excerpt
# ai-research-reproduction
## Purpose
Guide README-first deep learning reproduction toward the smallest trustworthy
run with auditable evidence. Preserve documented meaning; record assumptions,
deviations and blockers instead of changing semantics to manufacture success.
Load specialized references only for a concrete uncertainty.
## Fast Path
For a routine bounded run, keep the control path short:
1. Read the target README and only the target test/config/source needed to understand the documented command.