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.
git clone --depth 1 https://github.com/bakabaka91/claude-batonWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/commands/bakabaka91/claude-baton/memo-resume)<a href="https://agentmods.dev/commands/bakabaka91/claude-baton/memo-resume"><img src="https://agentmods.dev/badge/commands/bakabaka91/claude-baton/memo-resume/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/commands/bakabaka91/claude-baton/memo-resume"><img src="https://agentmods.dev/badge/commands/bakabaka91/claude-baton/memo-resume.svg" alt="Reviewed on agentmods" width="80" height="20"></a>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 5.1 | $0.00000 | $0.01011 |
| Opus 5 | $0.00000 | $0.00505 |
| Sonnet 5 | $0.00000 | $0.00202 |
| Haiku 4.5 | $0.00000 | $0.00101 |
Grade A, and why
memo-resume 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 11d 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.
How it starts
The opening of the file, as written. The whole thing — 92 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Restore context from last checkpoint at session start.
Important: Run all git commands exactly as written below — do NOT add -C flags or path arguments. The working directory is already the project root.
Steps
-
Detect the project from the current working directory.
-
Call the
list_checkpointsMCP tool with today's date. If multiple checkpoints exist, show them and note which is latest. -
Call the
get_checkpointMCP tool (latest by default, or by ID if the user specified one via $ARGUMENTS).
3.5. Staleness check: If a checkpoint was found, compute the time delta between the checkpoint's created_at and now:
- More than 7 days old: display
"⚠️ WARNING: This checkpoint is [N] days old. Project state may have changed significantly." - More than 24 hours but 7 days or less: display
"📌 Note: This checkpoint is [N] hours/days old." - 24 hours or less: no staleness note
3.75. Cold-start fallback: If NO checkpoint exists for this project:
- Run
git log --oneline -20,git diff --stat HEAD~5..HEAD,git branch --show-current,git status --short - Present a "🆕 Cold Start Briefing" with recent commits, file changes, and uncommitted work
- End with: "No checkpoint to resume from, but here is the project state from git. What would you like to work on?"
- Skip remaining steps (4-8)
-
Capture current git state by running these bash commands:
git branch --show-currentgit status --shortgit log --oneline -10git diff --stat main...HEAD
-
If the checkpoint has a
plan_referencefield (e.g. "docs/v2-plan.md Phase 2 Step 3"), read the referenced file and extract the relevant section. Parse the plan_reference to get the file path (everything before the section identifier like "Phase" or "Step") and the section name. -
Diff intelligence -- compare checkpoint state to current state:
- If checkpoint has a
git_snapshot, extract the top commit hash and rungit diff --stat <hash>..HEADto see what files changed since checkpoint - Compare checkpoint
uncommitted_filesvs currentgit status --short-- note any files that were added, removed, or committed since checkpoint - If
package.jsonappears in changed files, rungit diff <hash>..HEAD -- package.jsonto check for dependency changes - Summarize as: "X files modified, Y new commits, Z dependency changes" (or "No changes since checkpoint" if clean)
- If checkpoint has a
-
Present the structured handover briefing:
## 🔄 Session Resume — [Project Name] — [DATE]
📍 Resuming from checkpoint: [timestamp]
🏷️ Source: [auto/manual] checkpoint
[Staleness warning if applicable]
### 🌿 Branch
`[checkpoint branch]` — currently on `[current branch]`
[Flag if they differ with ⚠️]
### 📍 Where We Left Off
[current_state from checkpoint]
### 🏗️ What Was Built (Last Session)
[what_was_built from checkpoint]
### 🧭 Decisions Made
[decisions from checkpoint -- or "None recorded"]
### 🚧 Blockers
[blockers from checkpoint -- or "✅ None"]
[If blockers exist, surface them prominently with ⚠️]
### 📋 Active Plan Context
[If plan_reference exists: show the reference, then the extracted section content from step 5]
[If no plan_reference: "No active plan"]
### 🔀 Changes Since Checkpoint
[Diff intelligence summary from step 6]
[Files changed, new commits, dependency changes]
### 🎯 Next Steps
Present as numbered action items:
1. [first action from next_steps]
2. [second action if applicable]
...
### 📝 Uncommitted Work
[checkpoint uncommitted_files vs current git status]
[Flag if they differ with ⚠️ -- someone may have stashed or committed manually]
### 📜 Commits From Last Session
[git_snapshot from checkpoint -- these are the commits made during that session]
[If empty: "No commits were made in the last session"]
### 🆕 Git Activity Since Checkpoint
[Any NEW commits since the checkpoint timestamp that were NOT in git_snapshot]
[If none: "No new commits since checkpoint"]
- End with: "✅ Ready to continue. Want me to start with step 1, or are we doing something else?" (referring to the numbered action items above)
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.
- 11d ago First seen · 92 lines · 0 tokens per session scan A fe47ae7ced75
memo-resume is a command published in the GitHub repository bakabaka91/claude-baton (25 stars, last pushed 5mo ago), licensed MIT. It costs nothing until one of its globs matches a file; then it loads 1,011 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 2026-09-01.
Other commands, from other repositories
undo-commit
Soft-undo the last commit. Restores its changes as staged, leaves working tree intact. Refuses if HEAD is already pushed to a tracked remote.
session-report
Capture what changed this session and why, scoped to the current branch. Read by ship verbs when synthesizing the commit message; deleted after a successful commit.
commit
Stage, commit, and optionally ship further. Pass an escalation token (push, pr, merge, squash, squash merge) to skip the prompt. With no token, commits then asks how far to ship. Delegates message format to the atomic-git-discipline skill.
catchup
Restore context after /clear by summarizing recent work and project state.
refine-title.en
Rewrite an Issue or PR title in Conventional Commits format.
afcl
Close feature [agent] [--adopt] - merges branch, cleans up, optionally adopts from losers (shortcut for feature-close).