Skills Explorer Add Skills

app-platform-router

An agent skill by digitalocean-labs, from digitalocean-labs/do-app-platform-skills. Tags: app-platform, deployment, digitalocean, routing, troubleshooting.

What it does

Routes DigitalOcean App Platform tasks to specialized sub-skills. Use when working with App Platform deployments, migrations, database configuration, networking, or troubleshooting.

Install

With the skills CLI, which installs into Claude Code, Codex, Cursor and other agents:

npx skills add digitalocean-labs/do-app-platform-skills --skill app-platform-router

Or copy the skill folder into Claude Code's skills directory by hand (~/.claude/skills for every project, or .claude/skills inside one):

git clone --depth 1 https://github.com/digitalocean-labs/do-app-platform-skills
cp -r do-app-platform-skills ~/.claude/skills/app-platform-router

Safety box score

Grade D, elevated risk: 46 of 100 risk points, rated 2026-10-20. The grade rates what the skill and its scripts can reach on the machine of whoever installs it, across eight categories from shell execution to secrets access. How the score works.

This repository is mostly a router plus docs, but it also ships executable validation/maintenance scripts and a scheduled GitHub Action that collects analytics and pushes commits back to the repo. The main risks are bounded code execution/network access, live-container troubleshooting that can expose app secrets, and recursive cleanup or automation that mutates the checkout without a human check.

Category levels, from 0 (none) to 3 (high)
CategoryLevelWhy
Code & shell execution2 (moderate)The repo ships runnable entrypoints and shell targets (`Makefile`, `scripts/validate_skills.py`, `.github/workflows/collect-analytics.yml`) that execute Python, `pytest`, and `pip install` commands, so it is not read-only content.
Network access2 (moderate)The analytics workflow reaches out to PyPI and GitHub (`pip install requests PyGithub pandas`, `GITHUB_TOKEN: ${{ secrets.ANALYTICS_PAT }}`), and the troubleshooting patterns use live app connections/log retrieval.
Filesystem reach2 (moderate)The `clean` target recursively deletes workspace artifacts (`rm -rf htmlcov/`, `.pytest_cache/`, `.coverage`, and `find . -type d -name __pycache__ -exec rm -rf {} +`), so the skill can remove a lot of local files inside the checkout.
Secrets & credentials2 (moderate)The troubleshooting docs intentionally exec into running containers and inspect environment variables (`app.exec("env | grep DATABASE")`, `psql $DATABASE_URL`), and the deployment/analytics docs wire GitHub Secrets into automation.
Privilege & persistence0 (none)I found no `sudo`, service registration, shell-startup edits, or global OS changes; the repo only asks for local package installs and GitHub Action permissions.
Instruction hijack surface0 (none)No file tells the agent to ignore its rules or hide behavior; the strongest wording is routing guidance like `Trust the AI`, which is advisory rather than prompt injection.
Irreversible actions2 (moderate)The skill suite includes a scheduled workflow that auto-commits/pushes analytics and a router that chains into deployment/troubleshooting paths, so it can drive nontrivial actions without a human in the loop.
Transparency2 (moderate)Several key files could not be read in this pass (`scripts/collect_analytics.py`, `shared/credential-patterns.md`, multiple sub-skill `SKILL.md` files), and the analytics workflow also downloads unpinned PyPI packages at runtime.

Before installing

Rated 2026-10-20 by an automated review of 71 files from the skill's repository.

Source

Repository
digitalocean-labs/do-app-platform-skills (all skills from this repository)
Path
SKILL.md
Branch
main
Collection
do-app-platform
Updated
2026-09-21