principle-sequence-verifiable-units
An agent skill by cursor, from cursor/plugins. Tags: delivery, planning, testing, validation, workflow.
What it does
Apply to multi-step work (sweeps, migrations, runs of similar edits) and to how you stack commits and PRs. Break work into small units that each end in a verifiable state, check each before the next, and order delivery so the sequence proves itself to a reviewer.
Install
With the skills CLI, which installs into Claude Code, Codex, Cursor and other agents:
npx skills add cursor/plugins --skill principle-sequence-verifiable-unitsOr 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/cursor/plugins
cp -r plugins/pstack/skills/principle-sequence-verifiable-units ~/.claude/skills/principle-sequence-verifiable-unitsSafety box score
Grade A, minimal risk: 0 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 skill is documentation-only: it provides process guidance for sequencing work into small, verifiable steps and for ordering commits/PRs, but ships no scripts or code. The main risk is operational rather than technical—encouraging extra verification and disciplined sequencing—but it does not appear to touch files, networks, secrets, or system settings.
| Category | Level | Why |
|---|---|---|
| Code & shell execution | 0 (none) | `SKILL.md` only describes workflow principles like "Run it anyway"; it ships no shell scripts, code, or evaluators. |
| Network access | 0 (none) | There is no download, upload, API call, or remote-host interaction in the skill files. |
| Filesystem reach | 0 (none) | The file does not instruct writing or deleting outside the project, and contains no path-manipulating commands. |
| Secrets & credentials | 0 (none) | No credential stores, tokens, passwords, `.env` files, or login flows are mentioned. |
| Privilege & persistence | 0 (none) | The skill does not request sudo, global installs, services, or OS/shell configuration changes. |
| Instruction hijack surface | 0 (none) | There are no hidden instructions, prompt-injection cues, or directives to bypass confirmations or follow external content. |
| Irreversible actions | 0 (none) | It advises how to sequence work and checks, but does not itself perform actions like deploy, publish, send, or delete. |
| Transparency | 0 (none) | All shipped content is plain markdown; there are no binaries, base64 blobs, or uninspectable installers. |
Before installing
- No special install check is needed; this is a documentation-only skill.
- If you plan to use it in a workflow, read `SKILL.md` alongside the referenced sibling principles (`prove-it-works`, `build-the-lever`) so the sequence is consistent.
Rated 2026-10-20 by an automated review of 1 file from the skill's repository.
Source
- Repository
- cursor/plugins (all skills from this repository)
- Path
- pstack/skills/principle-sequence-verifiable-units/SKILL.md
- Branch
- main
- Collection
- cursor
- Updated
- 2026-09-23
Related skills
- create-verification-skill — Generate a project-local verification skill that drives your app the way a user does — any language, framework, or platform.
- maintain-verification-skill — Periodic pass that keeps a project's verification skill and feature map honest: parallel source readers per feature, one live session driving every feature.
- principle-outcome-oriented-execution — Apply during planned rewrites and migrations with explicit phase boundaries.
- tdd — Test-driven development. Use when the user wants to build features or fix bugs test-first, mentions "red-green-refactor", or wants integration tests.
- blast-radius — Find what a change could break somewhere else before it ships, beyond the diff.
- check-compiler-errors — Run compile and type-check commands and report failures