checkpoint

A command that saves the current conversation’s working state in a .memory/ folder. It records tasks, decisions, and unanswered questions in separate files.

In plain words
What is it for?
Use it to checkpoint progress during a coding session. It helps maintain todo lists, decision records, and open questions.
Why use it?
It reduces the risk of losing project context between sessions. The agent can preserve what remains to do and what has already been decided.

Command

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/backchainai/backchain-plugins/checkpoint
Clone the repo
git clone --depth 1 https://github.com/backchainai/backchain-plugins
Per session 0 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 520 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 $0.00000 $0.00520
Opus 5 $0.00000 $0.00260
Sonnet 5 $0.00000 $0.00104
Haiku 4.5 $0.00000 $0.00052

Measured yesterday against content hash 447fe0ee72ed, 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 yesterday.

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.

engram/skills/working/commands/checkpoint.md · 70 lines

How it starts

The opening of the file, as written. The whole thing — 70 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Command: checkpoint

Create or update .memory/ with current conversation state.

Trigger

/engram:working checkpoint or natural language: "checkpoint our progress", "save working memory".

Behavior

  1. Determine the current working directory.
  2. Check whether .memory/ exists; create it if not.
  3. For each template file (todos.md, decisions.md, questions.md):
    • If the file exists: read current contents, append new items, update the timestamp.
    • If it does not exist: create from the template in ${CLAUDE_SKILL_DIR}/templates/.
  4. Analyze the current conversation to extract:
    • Todos: action items, next steps, agreed-upon tasks.
    • Decisions: explicit choices with rationale, trade-off resolutions.
    • Questions: open issues, deferred decisions, knowledge gaps.
  5. Update each file's Last Updated timestamp using date -Iseconds.
  6. Confirm with a summary.

Extraction guidelines

Todos

Look for:

  • Explicit action items ("we need to…", "next step is…").
  • Agreed-upon tasks ("let's do X").
  • Follow-up items ("we should also…").

Categorize as: In Progress, Blocked, Completed, Backlog.

Decisions

Look for:

  • Explicit choices ("we decided to use X").
  • Trade-off resolutions ("X over Y because…").
  • Architecture / design selections.
  • Tool / library selections.

Capture: context, decision, rationale, alternatives considered.

Questions

Look for:

  • Unresolved discussions ("we still need to figure out…").
  • Deferred decisions ("let's decide later…").
  • Knowledge gaps ("I'm not sure about…").
  • External dependencies ("we need to ask…").

Capture: question, context, what it blocks, candidate answers.

Output format

Checkpointed working memory (.memory/):
- {N} todos ({new} new, {existing} existing)
- {N} decisions ({new} new, {existing} existing)
- {N} questions ({new} new, {existing} existing)

Constraints

  • Append; never overwrite. Preserving prior session context is the whole point.
  • Preserve template structure (headings, formatting).
  • Use RFC 3339 timestamps from date -Iseconds.
  • Keep items concise but include enough context to resume cold.
  • If no new items in a category, skip its file (do not create empty entries).

Read the full file on GitHub · 70 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.

  1. yesterday First seen · 70 lines · 0 tokens per session scan A 447fe0ee72ed

Subscribe to this mod's changes

checkpoint is a command published in the GitHub repository backchainai/backchain-plugins (4 stars, last pushed 27d ago), licensed Apache-2.0. It costs nothing until one of its globs matches a file; then it loads 520 tokens. 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 2026-08-31.