thermos

An agent skill by cursor, from cursor/plugins. Tags: analysis, code-review, multi-agent, review, security.

What it does

Launch both thermo-nuclear review subagents in parallel, then synthesize their findings. Use for thermos, double thermo review, or combined bug/security and code-quality branch audits.

Install

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

npx skills add cursor/plugins --skill thermos

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/thermos/skills/thermos ~/.claude/skills/thermos

Safety box score

Grade A, minimal risk: 0 of 100 risk points, rated 2026-10-16. 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 a review orchestrator: it reads the requested scope, collects diffs/context, and asks two background subagents to analyze the change. It does not include scripts or commands that directly modify the machine, network, files, or secrets; the main consideration is that it broadens what code/context the agent may inspect during review.

Category levels, from 0 (none) to 3 (high)
Category	Level	Why
Code & shell execution	0 (none)	No shell commands, code execution, or native scripts are present; the file only says to "Launch both subagents" and synthesize their results.
Network access	0 (none)	There is no downloading, uploading, or remote API call described in the skill text.
Filesystem reach	0 (none)	The workflow only says to "Gather the diff and any file/context excerpts" and does not instruct writes, deletes, or access outside the review scope.
Secrets & credentials	0 (none)	The skill does not ask for tokens, keys, passwords, dotfiles, or other credential stores.
Privilege & persistence	0 (none)	No sudo, installs, services, shell startup changes, or other privilege-bearing actions are mentioned.
Instruction hijack surface	0 (none)	The text does not tell the agent to ignore instructions, hide behavior, or follow external content; it stays within review coordination.
Irreversible actions	0 (none)	It coordinates internal review subagents, but it does not publish, deploy, message, pay, or delete anything.
Transparency	0 (none)	The only source file is readable markdown; there are no binaries, installers, or decoded-at-runtime blobs.

Before installing

Read `SKILL.md` to confirm it is only being used for review workflows.
If you use it on a sensitive branch, make sure the gathered diff/context excludes any secrets or unrelated private files.
Verify there are no additional scripts or hidden files in the skill package beyond the single markdown file shown here.

Rated 2026-10-16 by an automated review of 1 file from the skill's repository.

Source

Repository: cursor/plugins (all skills from this repository)
Path: thermos/skills/thermos/SKILL.md
Branch: main
Collection: cursor
Updated: 2026-09-23

interrogate — Use for "interrogate", "adversarial review", "multi-model review", "challenge this", "stress test this code", "find blind spots", or "tear this apart".
no-comments — Spawn Comment Sicko, fix accepted findings, and offer encodings for claimed constraints.
reflect — Spawn three parallel review subagents over the active transcript, surface learnings, and route each to a concrete edit on an existing skill.
thermo-nuclear-review — Comprehensive security and correctness audit of a branch's changes.
differential-review — Performs security-focused differential review of code changes.
trailmark-review-gate — Runs a Trailmark structural review gate over a branch, pull request, fix commit, release diff, or git ref range to detect new entrypoints, new tainted paths.