Skip to content
Skip to content
Guides / Agent skills10 min read

How to Make Your Agent Better at Workflow

A skills.sh shortlist for brainstorming, planning, TDD, review, and debugging, with a download link and install command for each skill.

Repath Khan
Critique

Critique

GUIDES / AGENT SKILLS

Better workflow

Brainstorm, plan, TDD, review.

critique.sh

Agent harness · BYOA · Composer 2.5

Direct answer

Install the Superpowers sequence from skills.sh: `using-superpowers`, `brainstorming`, `writing-plans`, `test-driven-development`, `requesting-code-review`, and `systematic-debugging`. Pair that with `find-skills` so the agent can fetch more skills instead of improvising. Skills shape the author loop. They do not replace an independent finish pass.

Most agents start coding in the first tool call. The expensive failure is not a syntax error. It is forty minutes of implementation against the wrong spec, followed by a test that documents the bug.

Workflow skills are gates. Brainstorming is a gate against coding. Planning is a gate against holding the entire feature in context. TDD is a gate against untested code. Review is a gate against the author grading itself. If those gates are optional, the agent will skip them whenever the prompt sounds urgent.

3.4M
find-skills installs
363.1K
brainstorming installs
259.2K
systematic-debugging
226.4K
test-driven-development

skills.sh is the public directory for Agent Skills: folders of instructions, and sometimes scripts, that a coding agent can load when the task matches. The package manager is npx skills. The discovery skill is `find-skills` from vercel-labs/skills (3.4M installs).

A skill is not a model upgrade. It is a procedure the agent is supposed to follow: which files to read, which checks to run, which output shape to return, and when to stop. That is why a mediocre prompt plus a good skill often beats a long chat with no procedure.

Default agent loop
Read the promptEdit filesMaybe run testsDeclare done
Skill-guided loop
BrainstormWrite the planRed-green TDDDispatch reviewDebug from evidence

`obra/superpowers` is the coherent pack for that loop. Matt Pocock’s listing is the other high-install workflow cluster: `implement` (556.7K), `code-review` (553.3K), and `diagnosing-bugs` (604.4K). Pick one pack as the default so skills do not contradict each other.

When to open which Superpowers skill
  1. 1
    Has anyone agreed what to build?
    brainstorming. No files, no scaffolding, no “quick spike” that becomes the architecture.
  2. 2
    Is the design approved but the work is still a blob?
    writing-plans. Demand file paths, verification commands, and tasks small enough to fail independently.
  3. 3
    Are you implementing behavior?
    test-driven-development. Watch the test fail. If it passes immediately, you tested the wrong thing.
  4. 4
    Did a unit of work just land?
    requesting-code-review. Give the reviewer SHAs and the spec, not the author transcript.

using-superpowers is the on-ramp. Without it, the rest of the pack is a folder the agent never opens. The other five skills are the actual gates. subagent-driven-development (204.9K) is the optional accelerator once the plan exists: one fresh subagent per task, then spec review, then quality review.

Skills, jobs, and download links
Install counts are from the skills.sh leaderboard on 15 September 2026. Use the count as a popularity filter, then read the skill before you install it.
The download link opens the skills.sh page. The install command below is the actual package fetch.
SkillUse it forInstallsDownload
using-superpowersLoad the Superpowers workflow so brainstorming, planning, TDD, and review fire as a sequence, not as optional extras.243.6KDownload on skills.sh
brainstormingRefuse to write code until the design is written down and the human has signed off.363.1KDownload on skills.sh
writing-plansBreak an approved design into small TDD tasks with file paths, commands, and expected output.248.7KDownload on skills.sh
test-driven-developmentWrite a failing test, watch it fail, then write the minimum code that passes. Delete code written before tests.226.4KDownload on skills.sh
requesting-code-reviewDispatch a reviewer subagent with SHAs and requirements, never the author session’s history.228.3KDownload on skills.sh
systematic-debuggingInvestigate failures with evidence instead of restarting the feature from a fresh prompt.259.2KDownload on skills.sh

Open the download page

Each card goes to the skills.sh listing. From there you can copy the install command or inspect the SKILL.md.

Bootstrap
using-superpowers
A bag of skills does nothing if the agent never opens them. This is the bootstrap skill for the rest of the pack. 243.6K installs. Source: obra/superpowers.
Design
brainstorming
Highest-install Superpowers skill. It is a hard gate: no scaffolding until the approach is explicit. 363.1K installs. Source: obra/superpowers.
Plan
writing-plans
Stops the agent from holding the whole feature in working memory. Each task should be small enough to review. 248.7K installs. Source: obra/superpowers.
Build
test-driven-development
The usual agent failure is implementation-first, then a test that mirrors the bug. This skill forbids that order. 226.4K installs. Source: obra/superpowers.
Review
requesting-code-review
A process skill, not an independent product. It still helps because the reviewer does not inherit the author’s story. 228.3K installs. Source: obra/superpowers.
Debug
systematic-debugging
When the tests go red, agents like to rewrite. This skill keeps them on the actual failing path. 259.2K installs. Source: obra/superpowers.
Install from skills.sh
find-skills first, then the domain pack. The agent reads SKILL.md after the files land.

discover

npx skills add https://github.com/vercel-labs/skills --skill find-skills
npx skills find tdd review workflow

Browse the leaderboard, then search by job rather than by brand.

using-superpowers

npx skills add https://github.com/obra/superpowers --skill using-superpowers

https://skills.sh/obra/superpowers/using-superpowers

brainstorming

npx skills add https://github.com/obra/superpowers --skill brainstorming

https://skills.sh/obra/superpowers/brainstorming

writing-plans

npx skills add https://github.com/obra/superpowers --skill writing-plans

https://skills.sh/obra/superpowers/writing-plans

test-driven-development

npx skills add https://github.com/obra/superpowers --skill test-driven-development

https://skills.sh/obra/superpowers/test-driven-development

requesting-code-review

npx skills add https://github.com/obra/superpowers --skill requesting-code-review

https://skills.sh/obra/superpowers/requesting-code-review

systematic-debugging

npx skills add https://github.com/obra/superpowers --skill systematic-debugging

https://skills.sh/obra/superpowers/systematic-debugging

`skill-creator` from Anthropic (381.5K installs) is the right tool when your team keeps re-explaining the same procedure. Do not start there. Start by installing a public skill that already encodes the loop. Create a private skill when you can name the failure: “agents skip the billing invariant,” not “we should have more skills.”

A workflow skill can still grade its own homework. Dispatching a reviewer subagent is better than no review. It is not the same as a controller-owned finish pass.

If Claude Code, Codex, Cursor, or OpenCode is already the author, keep it. Install the skills above, then run an independent finish pass with the critique sidecar when a unit of work is ready. If you want Critique to be the author, use CritiqueCode: it implements, then the controller forces review and verified repair before promotion. Those are different binaries. Do not mix the jobs.

Install the loop, then force the finish.
Workflow skills stop the agent from improvising. CritiqueCode is the author agent that still has to survive independent review before the work is promoted.
Primary sources
skills.sh directory
Open agent skills leaderboard. Counts in this series were read on 15 September 2026.
find-skills
Vercel’s discovery skill. 3.4M installs. Use it when you do not already know the package name.
using-superpowers (obra/superpowers)
Load the Superpowers workflow so brainstorming, planning, TDD, and review fire as a sequence, not as optional extras. 243.6K installs. npx skills add https://github.com/obra/superpowers --skill using-superpowers
brainstorming (obra/superpowers)
Refuse to write code until the design is written down and the human has signed off. 363.1K installs. npx skills add https://github.com/obra/superpowers --skill brainstorming
writing-plans (obra/superpowers)
Break an approved design into small TDD tasks with file paths, commands, and expected output. 248.7K installs. npx skills add https://github.com/obra/superpowers --skill writing-plans
test-driven-development (obra/superpowers)
Write a failing test, watch it fail, then write the minimum code that passes. Delete code written before tests. 226.4K installs. npx skills add https://github.com/obra/superpowers --skill test-driven-development
requesting-code-review (obra/superpowers)
Dispatch a reviewer subagent with SHAs and requirements, never the author session’s history. 228.3K installs. npx skills add https://github.com/obra/superpowers --skill requesting-code-review
systematic-debugging (obra/superpowers)
Investigate failures with evidence instead of restarting the feature from a fresh prompt. 259.2K installs. npx skills add https://github.com/obra/superpowers --skill systematic-debugging
CritiqueCode reference
Author agent that implements, then is forced through independent review and verified repair.
Critique CLI reference
Sidecar for an agent you already use. It reviews the working tree; it does not replace the author.
skill-creator
381.5K installs. Use when you have a repeatable procedure worth encoding, after public skills are in place.
subagent-driven-development
204.9K installs. One fresh subagent per plan task, with spec and quality review between tasks.
implement (mattpocock/skills)
556.7K installs. High-install alternative to Superpowers if you already live in that pack.
code-review (mattpocock/skills)
553.3K installs. Process review skill. Not a substitute for an independent finish pass.
Best code review skill for Claude, Hermes, Codex, OpenCode
Portable critique-review skill across author harnesses.
Independent reviewVerified repairReal repositoriesBuilt for developersLoved by agents