manage-renv-dependencies
About
This skill helps developers manage R package dependencies using renv to create reproducible environments. It handles initialization, snapshot/restore workflows, troubleshooting, and CI/CD integration for R projects. Use it when setting up new projects, restoring environments on different machines, or resolving dependency issues.
Quick Install
Claude Code
Recommendednpx skills add pjt222/agent-almanac -a claude-code/plugin add https://github.com/pjt222/agent-almanacgit clone https://github.com/pjt222/agent-almanac.git ~/.claude/skills/manage-renv-dependenciesCopy and paste this command in Claude Code to install this skill
Documentation
Manage renv Dependencies
Setup + maintain reproducible R pkg envs via renv.
Use When
- Init dep mgmt for new R project
- Add / update pkg deps
- Restore env on new machine
- Troubleshoot renv restore fails
- Integrate renv w/ CI/CD
In
- Req: R project dir
- Opt: Existing
renv.lock(for restore) - Opt: GitHub PAT for private pkgs
Do
Step 1: Init renv
renv::init()
Creates:
renv/dir (library, settings, activation script)renv.lock(dep snapshot)- Updates
.Rprofile→ activate renv on load
→ Project-local lib created. renv/ dir + renv.lock present. .Rprofile updated w/ activation.
If err: Hangs → check network. Fails on specific pkg → install manually first w/ install.packages() + rerun renv::init().
Step 2: Add Deps
Install pkgs as usual:
install.packages("dplyr")
renv::install("github-user/private-pkg")
Snapshot to record state:
renv::snapshot()
→ renv.lock updated w/ new pkgs + vers. renv::status() shows no out-of-sync.
If err: Snapshot reports validation errs → renv::dependencies() to check which pkgs actually used → renv::snapshot(force = TRUE) to bypass validation.
Step 3: Restore on Another Machine
renv::restore()
→ All pkgs installed at exact vers in renv.lock.
If err: Common issues: GitHub pkgs fail (set GITHUB_PAT in .Renviron), sys deps missing (install w/ apt-get on Linux), timeouts on large pkgs (options(timeout = 600) before restore), or binaries not avail (renv compiles from source → ensure build tools installed).
Step 4: Update Deps
# Update a specific package
renv::update("dplyr")
# Update all packages
renv::update()
# Snapshot after updates
renv::snapshot()
→ Target pkgs updated to latest compatible vers. renv.lock reflects new vers after snapshot.
If err: Update fails for specific pkg → install directly w/ renv::install("package@version") + snapshot.
Step 5: Check Status
renv::status()
→ "No issues found" or clear list of out-of-sync pkgs w/ actionable guidance.
If err: Status reports pkgs used but not recorded → renv::snapshot(). Recorded but not installed → renv::restore().
Step 6: Config .Rprofile for Conditional Activation
if (file.exists("renv/activate.R")) {
source("renv/activate.R")
}
Ensures project works even if renv not installed (CI envs, collaborators).
→ R sessions activate renv auto when starting in project dir. Sessions w/o renv still start w/o errs.
If err: .Rprofile causes errs → ensure file.exists() guard present. Never call source("renv/activate.R") unconditionally.
Step 7: Git Config
Track:
renv.lock # Always commit
renv/activate.R # Always commit
renv/settings.json # Always commit
.Rprofile # Commit (contains renv activation)
Ignore (already in renv's .gitignore):
renv/library/ # Machine-specific
renv/staging/ # Temporary
renv/cache/ # Machine-specific cache
→ renv.lock, renv/activate.R, renv/settings.json tracked by Git. Machine-specific dirs (renv/library/, renv/cache/) ignored.
If err: renv/library/ accidentally committed → remove w/ git rm -r --cached renv/library/ + add to .gitignore.
Step 8: CI/CD Integration
GitHub Actions → renv cache action:
- uses: r-lib/actions/setup-renv@v2
Automatically restores from renv.lock w/ caching.
→ CI pipeline restores pkgs from renv.lock w/ caching. Subsequent runs faster due to cached pkgs.
If err: CI restore fails → check renv.lock committed + up to date. Private GitHub pkgs → ensure GITHUB_PAT set as repo secret.
Check
-
renv::status()reports no issues -
renv.lockcommitted to VC -
renv::restore()works on clean checkout -
.Rprofileconditionally activates renv - CI/CD uses
renv.lockfor dep resolution
Traps
- Run
renv::init()in wrong dir: Always verifygetwd()first - Mix renv + sys library: After init, only use project library
- Forget to snapshot: After installing, always
renv::snapshot() --vanillaflag:Rscript --vanillaskips.Rprofile→ renv won't activate- Large lock files in diffs: Normal —
renv.lockdesigned to be diffable JSON - Bioconductor pkgs: Use
renv::install("bioc::PackageName")+ ensure BiocManager configured
→
create-r-package— includes renv initsetup-github-actions-ci— CI integration w/ renvsubmit-to-cran— dep mgmt for CRAN pkgs
GitHub Repository
Related Skills
llamaguard
OtherLlamaGuard is Meta's 7-8B parameter model for moderating LLM inputs and outputs across six safety categories like violence and hate speech. It offers 94-95% accuracy and can be deployed using vLLM, Hugging Face, or Amazon SageMaker. Use this skill to easily integrate content filtering and safety guardrails into your AI applications.
cost-optimization
OtherThis Claude Skill helps developers optimize cloud costs through resource rightsizing, tagging strategies, and spending analysis. It provides a framework for reducing cloud expenses and implementing cost governance across AWS, Azure, and GCP. Use it when you need to analyze infrastructure costs, right-size resources, or meet budget constraints.
quantizing-models-bitsandbytes
OtherThis skill quantizes LLMs to 8-bit or 4-bit precision using bitsandbytes, achieving 50-75% memory reduction with minimal accuracy loss. It's ideal for running larger models on limited GPU memory or accelerating inference, supporting formats like INT8, NF4, and FP4. The skill integrates with HuggingFace Transformers and enables QLoRA training and 8-bit optimizers.
dispatching-parallel-agents
OtherThis Claude Skill dispatches multiple agents to investigate and fix 3+ independent problems concurrently. It is designed for scenarios involving unrelated failures that can be resolved without shared state or dependencies. The core capability is parallel problem-solving, assigning one agent per independent problem domain to maximize efficiency.
