agentsclimarketplace

Checkpoint review

Skill nevitonsantana/adaptive-skills/skills/checkpoint-review

38 portable Agent Skills for disciplined, reviewable AI-assisted work across Codex, Claude Code, GitHub Copilot, and compatible agents.

Install
npx -y skills add nevitonsantana/adaptive-skills --skill checkpoint-review

Assembled from the repository path, not quoted from the project. Check it against their README if it does not work.

2 things to look at

  • no licenseNo license file was found in the repository. Code published without one is not open source by default, so using it at work is a question for whoever answers licensing questions where you are.
  • 1 stars1 stars. Stars are a popularity signal and not a quality one, but at this level it is likely that nobody has read this closely except its author, and you would be relying on your own review.

What its author says it does

Copied from the file, not written here

Insert a short review pause between rounds so work can continue, stop, or hand off with better discipline.

SKILL.md

2.5 KB, as published. Nobody here has run it

Overview

Use this skill when a round needs a short, useful pause before continuing, so the next move is chosen deliberately instead of by momentum.

When to Use

  • the session is getting long or heavy
  • the task changed shape mid-round
  • the team needs to decide whether to continue, stop, or hand off
  • a quick review is cheaper than carrying ambiguity forward

When NOT to Use

  • the task is still in a very small, obvious execution pass
  • the work needs full macro gate or escalation rather than a local checkpoint
  • the round has already closed cleanly with no meaningful uncertainty left

Core Moves

  1. State what changed in this round.
  2. State what is now known or proved.
  3. Name what still remains unresolved.
  4. Decide whether the next step belongs in the same round, a new round, or a handoff.
  5. Record the stop-or-continue rationale.

Optional Modules

  • Proof recap — list the checks or artifacts that justify the current state.
  • Boundary test — ask whether the next step crosses into a different ownership or operating context.
  • Continuation cap — add an explicit stop point when the team is tempted to keep going by inertia.

Activation Triggers

  • Use proof recap when the current state depends on validation that could be forgotten quickly.
  • Use boundary test when the next step may need another skill, owner, or review posture.
  • Use continuation cap when the round keeps growing even though the key objective is already met.

Expected Output

  • what changed
  • what is now proved
  • what remains open
  • continue / stop / handoff decision
  • rationale for that decision

Verification

  • The round outcome is clearer after the checkpoint than before it.
  • The continue-or-stop decision is explicit.
  • Proven work is separated from assumed work.
  • The checkpoint reduces ambiguity instead of adding ceremony.

Handoff Signals

  • The next step needs a different owner, skill, or operating context.
  • The round should stop and resume later with a fresh summary.
  • Macro review, escalation, or gate posture is now required.

Pairs Well With

  • workflow
  • task-chunking
  • handoff-summary

Anti-patterns

  • Treating a checkpoint as a long retrospective.
  • Continuing only because the team already has momentum.
  • Using a checkpoint to avoid making a real stop-or-continue decision.

Keep looking

Skills are one crate of 328,083. Ordering is by how many stacks a row turns up in, so the top of any crate is what has actually been picked rather than what has the most stars.