Getting it into your agent
It runs from inside its repository, so the clone comes first — what it calls does not travel with the file alone.
git clone --depth 1 https://github.com/hajibabaie/combinatorial-optimization-skillsnpx agentmods add skills/hajibabaie/combinatorial-optimization-skills/git-for-research-codeWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/skills/hajibabaie/combinatorial-optimization-skills/git-for-research-code)<a href="https://agentmods.dev/skills/hajibabaie/combinatorial-optimization-skills/git-for-research-code"><img src="https://agentmods.dev/badge/skills/hajibabaie/combinatorial-optimization-skills/git-for-research-code/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/skills/hajibabaie/combinatorial-optimization-skills/git-for-research-code"><img src="https://agentmods.dev/badge/skills/hajibabaie/combinatorial-optimization-skills/git-for-research-code.svg" alt="Reviewed on agentmods" width="80" height="20"></a>What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00117 | $0.10105 |
| Opus 5.5 | $0.00047 | $0.04042 |
| Sonnet 5.5 | $0.00023 | $0.02021 |
| Haiku 4.5 | $0.00012 | $0.01010 |
Grade A, and why
git-for-research-code scanned grade A with 1 finding against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 1mo ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Runs shell commandslowCapability
Expected in a hook, worth knowing in a rule or an instructions file.
out = subprocess.run( How it starts
The opening of the file, as written. The whole thing — 781 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Git for Research Code
You are an expert in version control for computational optimization research. This skill covers the solo-researcher git workflow: commit granularity tied to experiments, annotated tags that freeze the code state behind every paper table, ignore rules for solver logs and result artifacts, provenance stamping that links each result row to a commit hash, and branch strategies that make risky refactors safe. Use the pattern catalog below to make every reported number reproducible from a single hash.
Initial Assessment
Establish the following before recommending a workflow or writing tooling:
- Repository status. Does a repo exist already? If yes, run
git statusandgit count-objects -vHmentally through the user: is the worktree clean, and has the repo already been polluted with large result files or solver logs? - Team size. Solo PhD-style work, a 2-3 person lab project, or a larger team? Solo work permits rebase-based history cleanup and direct commits to
main; shared remotes require merge discipline and protected branches. - Artifact inventory. What does a run produce? Typical optimization artifacts: result CSV/parquet tables, solver logs (
.log), model files (.lp,.mps), solution files (.sol), checkpoints, convergence traces, figures. Each class needs an explicit track/ignore decision. - Instance data. Are benchmark instances small text files (TSPLIB-style, fine to commit), large binaries (need LFS or an external store plus checksums), or licensed data that must never enter a public repo?
- Experiment cadence. How often does the user change code and re-run? Daily parameter sweeps need a low-friction stamp-and-run loop; monthly campaign-style runs justify heavier manifests.
- Paper pipeline. Which deliverables consume results - LaTeX tables, figures, a results section? Every camera-ready number should trace back to a tag.
- Reproducibility horizon. Will results need re-running at revision time (6-12 months later)? If yes, environment capture (Python and package versions) must ride along with the git hash.
- Solver licensing. Gurobi license files (
gurobi.lic), WLS credentials, and cluster SSH keys must be ignored and scanned for before any push to a public remote. - Remote and backup. Is there a remote (GitHub/GitLab/institute server)? A repo with no remote is one disk failure away from losing the entire provenance chain.
- Existing damage. If history already contains multi-GB results or secrets, plan a history rewrite (
git filter-repo) before adopting the clean workflow, not after. - Tooling constraints. Python version, OS (path separators in hooks), whether the user can install the
pre-commitframework or needs plain.git/hooksscripts.
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 1mo ago First seen · 781 lines · 117 tokens per session scan A 97f88f3541f9
git-for-research-code is a skill published in the GitHub repository hajibabaie/combinatorial-optimization-skills (14 stars, last pushed 4mo ago), licensed MIT. It adds 117 tokens to every session and 10,105 once invoked, about $0.0005 per session on Opus 5.5. A static security scan graded it A with 1 finding (runs shell commands). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.
Other skills, from other repositories
release
CONTRIBUTOR TOOL - Cut a plugin release: bump plugin.json version, finalize CHANGELOG, update README if needed, gate on make ci, commit, tag vX.Y.Z, and create the GitHub release. Use when shipping a new plugin version. NOT distributed.
pr-review
Address feedback left on a GitHub pull request: fetch unresolved review threads, make agreed Elixir/Phoenix code fixes, reply, and resolve. Use for a PR URL/number or reviewer comments. NOT for pre-PR review, findings triage, or CI monitoring.
phx-pr-review
Address feedback left on a GitHub pull request: fetch unresolved review threads, make agreed Elixir/Phoenix code fixes, reply, and resolve. Use for a PR URL/number or reviewer comments. NOT for pre-PR review, findings triage, or CI monitoring.
craft-plugin-release
Releasing Craft CMS plugins — tagging, Packagist propagation, GitHub releases, branch promotion, shared-library ordering, history rewrites. ALWAYS load when cutting, preparing, verifying, or debugging a plugin release: bumping a version, dating a changelog, creating or moving a git tag, or checking what Packagist…
commit-push-pr
Full git workflow - creates branch, commits, pushes, and creates or updates a PR with summary and test plan.
commit
Create a single well-crafted git commit from current changes. Analyzes diff, follows repo's commit style, and writes a concise "why not what" message.