verify-this

An agent skill by cursor, from cursor/plugins. Tags: analysis, documentation, research, testing, validation.

What it does

Verify a claim with fresh local evidence: restate it falsifiably, capture baseline and treatment, compare artifacts, and return VERIFIED, NOT VERIFIED, or INCONCLUSIVE.

Install

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

npx skills add cursor/plugins --skill verify-this

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/cursor/plugins
cp -r plugins/cursor-team-kit/skills/verify-this ~/.claude/skills/verify-this

Safety box score

Grade A, minimal risk: 4 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 is a documentation-only skill for verifying a claim by running local repros/tests and comparing before/after artifacts. It does not ship code or networked tooling, and the main risk is limited local data capture if you choose to save screenshots, transcripts, HTTP bodies, or heap snapshots.

Category levels, from 0 (none) to 3 (high)
Category	Level	Why
Code & shell execution	0 (none)	No scripts or code are shipped; `SKILL.md` only describes a workflow and output format.
Network access	0 (none)	The skill does not instruct downloads or remote uploads; it only mentions local API/HTTP comparisons as evidence types.
Filesystem reach	1 (low)	It may write comparison artifacts under `/tmp/verify-this/<claim-slug>/` and can store screenshots, transcripts, HTTP bodies, or heap snapshots there.
Secrets & credentials	0 (none)	It explicitly says to keep sensitive data minimal and does not ask for passwords, tokens, or key files.
Privilege & persistence	0 (none)	No sudo, service installation, shell startup edits, or other privilege changes are requested.
Instruction hijack surface	0 (none)	There is no instruction to ignore safeguards, hide behavior, or follow fetched instructions; the text is procedural and bounded.
Irreversible actions	0 (none)	The skill only asks for a verdict (`VERIFIED`, `NOT VERIFIED`, or `INCONCLUSIVE`) and does not itself publish, deploy, pay, or delete anything.
Transparency	0 (none)	There are no binaries, obfuscated blobs, installers, or unpinned remote code paths in the provided files.

Before installing

Read `SKILL.md` and confirm you are okay with the agent collecting local before/after artifacts for a claim.
If the claim may expose secrets or private data, plan to keep evidence inline rather than writing screenshots, HTTP bodies, or heap snapshots to disk.
Verify you want a verification-only workflow; this skill will not fix code or make remote changes on its own.

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: cursor-team-kit/skills/verify-this/SKILL.md
Branch: main
Collection: cursor
Updated: 2026-09-23

deep-research — Generate format-controlled research reports with evidence tracking, citations, source governance, and multi-pass synthesis.
blast-radius — Find what a change could break somewhere else before it ships, beyond the diff.
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-prove-it-works — Apply after completing a task, before declaring done.
recall — Reconstruct your recent working context from your own chat history, live state, and the shared record (user reports, prior fixes, incidents).
audit-prep-assistant — Prepares codebases for security review using Trail of Bits' checklist.