resume

resume is a skill for Claude Code, Codex from tianhanz/ears. It costs 40 tokens per session (1,424 once invoked), scanned A, original, MIT.

A way to restore work from a checkpoint file, which records task state for another session, branch, or agent. It finds or loads the relevant checkpoint and verifies the recorded artifacts.

In plain words
What is it for?
Use it after a crash or context loss, when handing work to another person or agent, when starting from a branch checkpoint, or when explicitly resuming a saved task.
Why use it?
It helps continue interrupted work without reconstructing the task from memory or starting again.

Skill for Claude CodeCodex

Written for no agent in particular: nothing here depends on one. Also seen: positional $N argument.

Good fit Use it after a crash or context loss, when handing work to another person or agent, when starting from a branch checkpoint, or when explicitly resuming a saved task.

Compare 6 skills from other repositories ↓
Install with agentmods
npx agentmods add skills/tianhanz/ears/resume
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.

Any agent
npx skills add tianhanz/ears --skill resume
Clone the repo
git clone --depth 1 https://github.com/tianhanz/ears

Made for: Claude Code, Codex.

Wrote 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.

agentmods badge for resume

README.md
[![agentmods](https://agentmods.dev/badge/skills/tianhanz/ears/resume/github.svg)](https://agentmods.dev/skills/tianhanz/ears/resume)
Your own site
<a href="https://agentmods.dev/skills/tianhanz/ears/resume"><img src="https://agentmods.dev/badge/skills/tianhanz/ears/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.

agentmods 80×15 button for resume

Your own site · 80×15
<a href="https://agentmods.dev/skills/tianhanz/ears/resume"><img src="https://agentmods.dev/badge/skills/tianhanz/ears/resume.svg" alt="Reviewed on agentmods" width="80" height="20"></a>
Per session 40 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 1,424 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. A grade says what 26 rules found in the file — not that it is safe.
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 5.1 $0.00040 $0.01424
Opus 5 $0.00020 $0.00712
Sonnet 5 $0.00008 $0.00285
Haiku 4.5 $0.00004 $0.00142

Measured 8d ago against content hash 351f06e07827, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-08, from the pricing page.

Security

Grade A, and why

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 8d 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/resume/SKILL.md · 181 lines

How it starts

The opening of the file, as written. The whole thing — 181 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Resume Skill

Load and restore task state from a structured checkpoint file.

When to Use

  • Session start: After context loss or session restart
  • Crash recovery: Picking up after unexpected interruption
  • Task handoff: Another person/agent continuing work
  • Branch fork: Starting from another branch's checkpoint
  • Manual resume: User explicitly requests /resume [checkpoint-file]

Step 1 — Find checkpoint

If no argument provided: Find latest checkpoint on current branch:

BRANCH=$(git branch --show-current)
LATEST=$(ls -t .claude/checkpoints/${BRANCH}-*.yaml 2>/dev/null | head -1)

if [ -z "$LATEST" ]; then
  echo "No checkpoint found for branch: $BRANCH"
  echo "Available checkpoints:"
  ls -t .claude/checkpoints/*.yaml 2>/dev/null | head -10
  exit 1
fi

echo "Found checkpoint: $LATEST"
CHECKPOINT_FILE="$LATEST"

If argument provided: Use that checkpoint file or branch name:

ARG="$1"
if [ -f "$ARG" ]; then
  CHECKPOINT_FILE="$ARG"
elif ls -t .claude/checkpoints/${ARG}-*.yaml 2>/dev/null | head -1 > /dev/null; then
  CHECKPOINT_FILE=$(ls -t .claude/checkpoints/${ARG}-*.yaml | head -1)
else
  echo "Checkpoint not found: $ARG"
  exit 1
fi

Step 2 — Load and display checkpoint

import yaml
import os

checkpoint_file = os.environ.get('CHECKPOINT_FILE', '.claude/checkpoints/latest.yaml')

with open(checkpoint_file, 'r') as f:
    checkpoint = yaml.safe_load(f)

print("=" * 60)
print(f"CHECKPOINT RESUME: {checkpoint_file}")
print(f"Created: {checkpoint['created_at']}")
print(f"Branch: {checkpoint['branch']}")
print("=" * 60)
print()

# Phase
phase = checkpoint.get('phase', {})
print("PHASE")
print(f"  Current: {phase.get('current', '?')}")
print(f"  Completed: {', '.join(phase.get('completed', []))}")
print(f"  Next: {', '.join(phase.get('next', []))}")
print()

# Parameters
params = checkpoint.get('state', {}).get('parameters', [])
if params:
    print(f"PARAMETERS ({len(params)} extracted)")
    for p in params:
        unit = p.get('unit', '')
        print(f"  {p['name']} = {p['value']} {unit} [{p.get('confidence','?')}] -- {p.get('source','?')}")
    print()

# Experiments
experiments = checkpoint.get('state', {}).get('experiments', [])
if experiments:
    print(f"EXPERIMENTS ({len(experiments)} total)")
    for exp in experiments:
        progress = f" {int(exp['progress']*100)}%" if 'progress' in exp else ""
        reason = f" -- {exp['reason']}" if 'reason' in exp else ""
        print(f"  {exp['name']} [{exp['status']}]{progress}{reason}")
    print()

# Decisions
decisions = checkpoint.get('state', {}).get('decisions', [])
if decisions:
    print(f"DECISIONS ({len(decisions)} made)")
    for dec in decisions:
        print(f"  {dec.get('question','?')} -> {dec.get('decision','?')} [{dec.get('confidence','?')}]")
    print()

# Blockers
blockers = checkpoint.get('blockers', [])
if blockers:
    print(f"BLOCKERS ({len(blockers)} open)")
    for b in blockers:
        assumption = f" (assuming: {b['assumption']})" if 'assumption' in b else ""
        print(f"  [{b.get('impact','?')}] {b['question']}{assumption}")
    print()

# Artifacts — verify they exist
artifacts = checkpoint.get('artifacts', [])
if artifacts:
    print(f"ARTIFACTS ({len(artifacts)} tracked)")
    missing = 0
    for a in artifacts:
        path = a['path'].replace('~', os.path.expanduser('~'))
        exists = os.path.exists(path)
        status = "OK" if exists else "MISSING"
        if not exists:
            missing += 1
        print(f"  [{status}] {a['path']} ({a.get('role','?')})")
    if missing > 0:
        print(f"\n  WARNING: {missing} artifact(s) missing!")
    print()

# Resume instructions
print("=" * 60)
print("RESUME INSTRUCTIONS")
print("=" * 60)
print()
print(checkpoint.get('resume_instructions', '(no instructions)'))
print()
print("=" * 60)

Read the full file on GitHub · 181 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. 8d ago First seen · 181 lines · 40 tokens per session scan A 351f06e07827

Subscribe to this mod's changes

resume is a skill published in the GitHub repository tianhanz/ears (5 stars, last pushed 4mo ago), licensed MIT. It adds 40 tokens to every session and 1,424 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 2026-08-31.

Related

Other skills, from other repositories

testing-api-authentication-weaknesses

Tests API authentication mechanisms for weaknesses including broken token validation, missing authentication on endpoints, weak password policies, credential stuffing susceptibility, token leakage in URLs or logs, and session management flaws. The tester evaluates JWT implementation, API key handling, OAuth flows, and…

xalgorix/xalgorix · 102 tokens

Effective Memory

The essential habits for an AI agent with memory — session bookends, learning triggers, verification, safety, and the operational discipline that turns raw recall into compounding intelligence. Pinned, always-injected.

plur-ai/plur · 44 tokens

s-compact

Session handoff for the next session — write-side pair of /s-continue. Triggers on "s-compact", "session handoff", "handoff prompt", "hand off", "wrap up the session", "prep for next session". Distills what /s-continue cannot recover: subagent findings, tool-output numbers, and decisions that never entered the…

ww-w-ai/super-token-saver · 99 tokens

project-memory

Generate a project-specific context file from a brief so an AI assistant remembers your editorial constraints, voice, audience, and quality bar across sessions.

ur-grue/autopunk-media-skills · 31 tokens

start

Session initialization and lifecycle management: bootstraps session context, organizes files, generates CLAUDE.md, manages soul purpose lifecycle with completion protocol and active context harvesting. Use when user says /start, /init, bootstrap session, initialize session, or organize project.

anombyte93/atlas-session-lifecycle · 55 tokens

test-spec-gen

Universal test specification generator that explores codebases, researches best practices, and generates comprehensive test specs via multi-agent orchestration. Outputs Hermes-style test specification documents with TC-XXX formatting, area segmentation, and optional Trello card conversion.

anombyte93/atlas-session-lifecycle · 50 tokens