checkpoint

A mechanism for saving a bounded record of work so a long task can resume after a context reset. The record includes completed steps, produced files, pending work, and the exact next action.

In plain words
What is it for?
Use it to record verified progress at milestones, preserve the state of unfinished implementation work, and resume a task accurately.
Why use it?
It prevents progress and decisions from being lost when a session becomes too long or needs to restart.

Skill for Claude CodeCodex

Part of the kernel plugin — 28 skills, 13 agents, 9 hooks shipped together

Install

Getting it into your agent

One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.

agentmods
npx agentmods add skills/ariaxhan/kernel-claude/checkpoint
Any agent
npx skills add ariaxhan/kernel-claude --skill checkpoint
Clone the repo
git clone --depth 1 https://github.com/ariaxhan/kernel-claude

Made for: Claude Code, Codex.

Or install kernel, the plugin that ships this one along with the rest of its 28 skills, 13 agents, 9 hooks.

Per session 34 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 689 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. Scan, not verified.
Origin original No closer match found in the catalogue.
Token cost

What it costs to keep this loaded

Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.

ModelPer sessionOnce invoked
Fable 5 $0.00034 $0.00689
Opus 5 $0.00017 $0.00345
Sonnet 5 $0.00007 $0.00138
Haiku 4.5 $0.00003 $0.00069

Measured 3d ago against content hash ac40162296b1, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

checkpoint scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 3d ago.

A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.

Nothing flagged

None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.

skills/checkpoint/SKILL.md · 75 lines

What it actually says

When to fire (EXP-L21: attention efficiency decays to ~11% late-session; a reset from a checkpoint restores the flat ~50-70k load-bearing line):

  • a long task crosses a natural boundary (a commit lands, a phase completes)
  • context is about to compact, or the session is degrading (anchor-drift, I0.13)
  • the manifest's execution.checkpoints list says so

Checkpoint vs handoff: same task + same session continuation → checkpoint. Ending the session / transferring to a future session with decisions and policy → handoff.

  1. Write _meta/checkpoints/{task}-{timestamp}.json per schemas/kernel.checkpoint.v1.schema.json:

    • task: goal + handoff_ref/contract_ref if they exist
    • steps_completed: each with EVIDENCE (commit sha, test output, command result), never intent. Receipts describe intent; files describe reality.
    • current_outputs: files produced so far
    • provenance.dirty_tree_sha256: when dirty, sha256 of git diff --binary HEAD
    • pending_steps: what remains, in order
    • resume: position (exact place in the work), entrypoint (skill to re-enter), next_operation (the LITERAL first action — "retarget tests/run-tests.sh lines ~1250 registration tests", not "continue the migration")
    • context (optional): required selectors + budget if the resume should be bounded

    Quote every sha/commit string (all-digit shas parse as integers and fail validation).

  2. Validate — MANDATORY:

    "${CLAUDE_PLUGIN_ROOT:-.}/orchestration/manifest/kernel-manifest" validate _meta/checkpoints/{file}.json
    
  3. Record:

    agentdb write-end '{"skill":"checkpoint","saved_to":"_meta/checkpoints/{file}.json","position":"..."}'
    

Resume path: /kernel:ingest discovers the newest manifest automatically (kernel-manifest latest — checkpoints and handoffs share discovery), or accepts an explicit path.

<hard_stops>

  • steps_completed without evidence → not a checkpoint, a hope. Add evidence or move the step to pending_steps.
  • unvalidated json → does not exist. Validate before reporting done. </hard_stops>
Changes

What this file has done since we first saw it

Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.

  1. 3d ago First seen · 75 lines · 34 tokens per session scan A ac40162296b1

Subscribe to this mod's changes

checkpoint is a skill published in the GitHub repository ariaxhan/kernel-claude (12 stars, last pushed 3d ago), licensed MIT. It adds 34 tokens to every session and 689 once invoked, about $0.0002 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.

Related

Other skills, from other repositories

prolong

Recover and use durable coding-session history from PRO-LONG's local append-only log. Use on long-running coding tasks, after context compaction or session resume, when reconstructing prior decisions or tool results, or before repeating work that may already have been attempted.

alexisfox7/PRO-LONG · 55 tokens

lemmalog

Externalize working memory and logical state into the lemmalog Datalog engine (MCP). Use for ANY multi-step task where state should outlive one context window or span agents: long investigations, debugging sessions, audits, multi-agent searches, systematic explorations, planning with many interdependent constraints…

JordyZomer/lemmalog · 105 tokens

causal-memory

Causal memory for agents — install/setup the causal-memory MCP server, then record decisions/outcomes and recall them before acting. Trigger when the user asks to install or set up causal-memory/agent memory, when causal-memory MCP tools are available and the agent faces a non-trivial decision (architecture, debugging…

JingxuanC/causal-memory · 88 tokens

a0-create-plugin

Create, extend, or modify Agent Zero plugins. Follows strict full-stack conventions (usr/plugins, plugin.yaml, Store Gating, AgentContext, plugin settings). Use for UI hooks, API handlers, lifecycle extensions, or plugin settings UI.

Gen-Verse/PAST-Bench · 54 tokens

a0-review-plugin

Full audit of Agent Zero plugins in usr/plugins/. Reviews manifest validity, directory structure, code patterns (Store Gating, notifications, imports), security, and duplicate detection against the community index. Use when asked to review, audit, validate, or check an existing plugin before using or contributing it.

Gen-Verse/PAST-Bench · 64 tokens

create-skill

Wizard for creating new Agent Zero skills. Guides users through creating well-structured SKILL.md files. Use when users want to create custom skills.

Gen-Verse/PAST-Bench · 33 tokens