skill CX Codex

reflect · neolabhq — Claude Code skill: critique your agent's output

Q: How do I install reflect in Claude?

Run `npx skills add NeoLabHQ/context-engineering-kit --skill reflect -a claude-code`. The skill lands in `~/.claude/skills/reflect/SKILL.md`.

A skill that makes the agent reread its own answer and list the flaws before you accept the work.

Install into Claude Code · Skills CLI (Vercel) All methods and platforms ↓

npx skills add NeoLabHQ/context-engineering-kit --skill reflect -a claude-code

Open on GitHub

What reflect does

reflect ships in the context-engineering-kit collection. It runs the agent's last response through a checklist: completeness, quality, correctness, dependencies, factual claims, and verification of generated files. There are no scripts inside, just a set of steps and questions the agent works through.

Tasks get sorted by complexity first. A single-file edit or a plain question takes the short path. Multi-file changes, new features and architecture calls get the full pass, with a required confidence above 4.0 out of 5.0. Core system changes, security code and performance-critical sections raise that bar to 4.5 out of 5.0.

A separate block covers code. Before writing anything custom, the agent has to look for an existing library or service. Date handling, form validation, HTTP requests, state management — those problems already have packages, and the skill demands a check. Custom code stays justified for business logic that exists nowhere else.

Worth installing if your agent tends to ship on the first try and you keep catching missed edge cases. One caveat: the instruction tone is deliberately harsh. The agent takes the role of a fault-finding reviewer, not a cheerleader.

What it helps with

reread its own answer and spot missed requirements
check whether an edit breaks something that depends on it
filter out invented claims about performance and security
catch leaked paths and keys in generated files
stop writing a helper where a library already exists

Example prompts for the agent

reflect on your last response and list the problems you find
check this change against the dependency checklist before recommending a file removal
run a reflection pass on this code: is there a library instead of my own implementation

How to install reflect

1Skills CLI (Vercel)

npx skills add NeoLabHQ/context-engineering-kit --skill reflect -a claude-code

2Manually: put the files in place

~/.claude/skills/reflect/SKILL.md

Copy the whole skill folder; for one project use .claude/skills/reflect/

1Skills CLI (Vercel)

npx skills add NeoLabHQ/context-engineering-kit --skill reflect -a codex

2Manually: put the files in place

.agents/skills/reflect/SKILL.md

For all projects use ~/.agents/skills/

1Skills CLI (Vercel)

npx skills add NeoLabHQ/context-engineering-kit --skill reflect -a cursor

2Manually: put the files in place

.cursor/skills/reflect/SKILL.md

Cursor also reads .agents/skills/ and .claude/skills/

1GitHub CLI

gh skill install NeoLabHQ/context-engineering-kit reflect

2Manually: put the files in place

.github/skills/reflect/SKILL.md

1Skills CLI (Vercel)

npx skills add NeoLabHQ/context-engineering-kit --skill reflect -a gemini-cli

2Platform CLI

gemini skills install https://github.com/NeoLabHQ/context-engineering-kit --path skills/reflect

Manually: put the files in place

.devin/skills/reflect/SKILL.md

The legacy .windsurf/skills/ path still works

Skills CLI (Vercel)

npx skills add NeoLabHQ/context-engineering-kit --skill reflect -a opencode

Step by step, with commands for each platform: How to install skills in Claude Code: five ways

Questions about reflect

How do I install reflect in Claude?

Run npx skills add NeoLabHQ/context-engineering-kit --skill reflect -a claude-code. The skill lands in ~/.claude/skills/reflect/SKILL.md.

What does the reflect skill do?

It makes the agent review its own last response against a checklist: completeness, correctness, dependencies, factual claims, generated file verification. The agent then either produces a fix plan or confirms the work.

How does the skill pick the review depth?

By task complexity. Simple edits take the short check, multi-file and architecture work needs the full pass with confidence above 4.0 out of 5.0, and critical sections need above 4.5 out of 5.0.

reflect by other authors comes first.

reflect backnotprop Mines the current session transcript using three review subagents and turns discovered lessons into concrete skill edits. CX Codex +1 5,300installs reflect Alireza Rezvani Pauses active task execution to re-evaluate conversation direction, assumptions, and bias across five analytical dimensions. 320installs reflect haowjy Reviews delivered work before the final report and routes session knowledge into project files. Useful for catching overlooked needs while context is fresh. CX Codex +1 378installs mantis-reflect official google Extracts insights, failures, and false assumptions from Mantis agent transcripts, appending them to workspace/learnings.jsonl. CX Codex +1 1,500installs memory-reflect basicmachines-co A background skill that reviews recent sessions and daily notes, pulls out what matters and folds it into long-term memory. Runs on a schedule or on demand. CX Codex +1 746installs reflect-skill Alireza Rezvani Halts detail work and re-reads the whole conversation from the original goal, so you can see where the direction drifted. ★ 28,733repo stars caveman-review JuliusBrussee Turns code review comments into one line per finding: location, problem, fix. Useful when a diff or PR needs a fast pass without the filler. CX Codex +1 411,800installs thermo-nuclear-code-quality-review cursor Runs a strict maintainability audit of code changes, abstractions, large files, and branching complexity. Use it when a routine review needs deeper structural scrutiny. CX Codex +1 22,200installs simplify brianlovin Rewrites code you just wrote so it reads better: less nesting, fewer duplicates, clearer names. Behaviour stays exactly as it was. CX Codex +1 20,100installs

From the same set55

prompt-engineering Guides prompt design for agents, subagents, and LLM integrations. Helpful when building production templates, chain-of-thought prompts, or agent hooks. CX Codex +1 1,400installs agent-evaluation Framework for testing agents, prompts, and skills using multi-dimensional scoring rubrics and LLM-as-judge prompts. CX Codex +1 1,300installs memorize Transfers insights from agent reflections and critique sessions into CLAUDE.md. Keeps your context file sharp by structuring lessons instead of bloating it with raw logs. CX Codex +1 1,300installs subagent-driven-development Runs an implementation plan through fresh subagents and puts code review between tasks, so context stays clean and issues surface early. CX Codex +1 1,300installs judge-with-debate The skill runs three independent judges who argue over a solution's score until they agree. Useful when a single pass feels too thin for a real verdict. CX Codex +1 1,300installs update-docs Updates project documentation based on uncommitted git changes or the latest commit. Keeps README files, docs/, and JSDoc comments synchronized with your codebase. CX Codex +1 1,300installs root-cause-tracing A skill that walks the agent backward through the call stack until it finds the original trigger, then fixes there instead of at the crash site. CX Codex +1 1,300installs critique The critique skill runs finished work past several AI judges: they review it independently, debate their findings, and agree on a shared list of issues. CX Codex +1 1,300installs