A repository handoff file that collects relevant project knowledge from multiple topic wikis. It can be exported, refreshed, checked, or imported, with privacy checks.

Use it to create or update a project knowledge checkpoint, verify one, or import one into a repository.

It keeps a project’s shared knowledge portable and easier to hand to another person or workspace. Privacy checks help catch information that should not be included.

checkpoint — Command by nvk

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 commands/nvk/llm-wiki/checkpoint

Clone the repo

git clone --depth 1 https://github.com/nvk/llm-wiki

Per session 24 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.

When invoked 1,700 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.

Model	Per session	Once invoked
Fable 5	$0.00024	$0.01700
Opus 5	$0.00012	$0.00850
Sonnet 5	$0.00005	$0.00340
Haiku 4.5	$0.00002	$0.00170

Measured 2d ago against content hash 9ab213e03d02, 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 2d 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.

claude-plugin/commands/checkpoint.md · 161 lines

How it starts

The opening of the file, as written. The whole thing — 161 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Your task

Manage the Project Knowledge Checkpoints Export: a portable, comprehensive project handoff synthesized from relevant compiled knowledge across topic wikis, with mandatory privacy checks. Read skills/wiki-manager/references/checkpoints.md before acting.

This is not a topic export, raw/session dump, or Git publication command. create and refresh are dry-run by default. --apply authorizes only the previewed file write—not commit, push, publication, release, or import.

Resolve and parse

Resolve HUB from $HOME/.config/llm-wiki/config.json using normal hub_path/fallback rules; read HUB/_index.md and HUB/wikis.json.
Parse create, refresh, verify, or import; natural “checkpoint/dump this project's wiki knowledge into the repo” means create.
For create, resolve --repo or git rev-parse --show-toplevel, symlinks, revision, remote, and authoritative visibility. Never guess visibility or write outside the resolved repository.
Resolve the lowercase slug from an explicit value, selected wiki Project, or repo name. Ask only if ambiguity changes scope.

Useful flags: --project <slug>, primary --wiki, repeatable --with <wiki>, --audience private|team|public, --include-archived, repeatable --allow-source <wiki:path>, repeatable --allow-finding <id>, required --override-reason "..." for overrides, and --apply.

There is no option to disable privacy scanning.

Create

Project packet and selection

Read bounded repo context: README, supplied WHY.md/BRIEF.md, ADRs, architecture/operations/security docs, manifests, and explicit seeds. Exclude .git, dependencies, builds, caches, env files, and arbitrary user folders.

Read active topic roots/indexes; select compiled articles and material Ideas, Projects, plans, and outputs by goal, aliases, decisions, and constraints. Follow one needed support hop. Archives require --include-archived; raw is not export content.

For each input record topic, wiki-relative path, SHA-256, inclusion reason, confidence, public|team|private|unknown access, and reuse note. Record material exclusions and gaps.

Read the full file on GitHub · 161 lines

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.

2d ago First seen · 161 lines · 24 tokens per session scan A 9ab213e03d02

Subscribe to this mod's changes

checkpoint is a command published in the GitHub repository nvk/llm-wiki (1,175 stars, last pushed 5d ago), licensed MIT. It adds 24 tokens to every session and 1,700 once invoked, about $0.0001 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.

Data and AI LLM Productivity agentic-ai agentic-skills agentic-workflow claude-code codex

checkpoint

Getting it into your agent

What it costs to keep this loaded

Grade A, and why

Nothing flagged

How it starts

Your task

Resolve and parse

Create

Project packet and selection

What this file has done since we first saw it

Other commands, from other repositories

implement

workflow-claude-commands

weather-orchestrator

project

review

research-verify