Skip to content
Agentmart
Submit a tool

agent-sync

Synchronize a shared repository after switching between Codex and Claude.

Skill by shabanakhtar

IndexedMIT
Write a review

Chalk Labs

Need help deploying agent-sync?

Chalk Labs sets up agents like this inside real teams: connected to your tools, tested on your data, with a human in the loop where it matters.

Book a call

Install

Works with Claude Code, Claude apps, Codex, Cursor, Gemini CLI, GitHub Copilot, VS Code, OpenCode, Goose, Windsurf. Read the repository before you give any tool access to your data.

npx agentmarket install skill/shabanakhtar-agent-sync

Asks which of your tools to add it to (Claude Code, Codex, Cursor, VS Code, Windsurf, Gemini CLI, Claude Desktop) and shows the change first. See https://agentmart.dev/cli.

Get the weekly diff

The weekly diff: one email a week with new agent tools, notable updates and grade changes, plus security alerts for tools you follow. Double opt-in, one-click unsubscribe.

We use your email only for this newsletter. Privacy

About

Synchronize a shared repository after switching between Codex and Claude. Invoke manually as `/agent-sync check` after returning from the other AI or `/agent-sync end` before handing work off.

Use cases: agent-building

No reviews yet. Installed it? Reviews come from verified installers only. How reviews work

Permissions

What this skill can do and what it asks for, from its SKILL.md. Checked Oct 25, 2026.

Tools
No tools found
Secrets
None asked for
Install scripts
None found

Network hosts it talks to 1

  • codeload.github.com · found in install command

Files it touches

  • ~/.claude/skills/agent-sync

What's inside

Skill name
agent-sync
Description
Synchronize a shared repository after switching between Codex and Claude. Invoke manually as `/agent-sync check` after returning from the other AI or `/agent-sync end` before handing work off.
License
MIT
Instructions
36 lines in SKILL.md · view on GitHub

Agent Skills spec issues

  • name: "agent-sync" does not match folder "claude"

Files (1)

  • SKILL.md

SKILL.md (first lines)

# Agent Sync

Use the project files as the source of truth. Do not rely on stale chat context.

## Modes

### `/agent-sync check`

Use after an agent switch and before repository-changing work when another AI may have worked since the last sync.

1. Read `AGENTS.md`, `CLAUDE.md` when present, `AGENT_HANDOFF.md`, and the relevant part of `PROJECT_PLAN.md`.
2. Inspect `git status`, recent commits, and diffs. Read files named under **Active Work** or open **Review Ledger** rows.
3. Review the other agent's proposed change or request. Record a concise result in its ledger row: `agree`, `needs change`, `implemented`, or `not applicable`, with the reviewed paths.
4. Update only the current agent's line in **Sync State**. Do not create artificial history entries.
5. Report the actual state succinctly. Do not implement new website work unless the user explicitly authorizes it.

Do not commit, push, stage unrelated files, or erase another agent's uncommitted work during `check`.

### `/agent-sync end`

Use before handing work to the other AI.

1. Run the same reconciliation checks as `check`; resolve or clearly leave open every relevant ledger row.
2. Update **Active Work**, **Next Step**, **Sync State**, and only the handoff details needed by the next agent.
3. Keep `AGENT_HANDOFF.md` compact. The proposing agent removes a resolved ledger row after the other AI's review has been consumed. Non-core resolved notes may be removed immediately; preserve user decisions, active blockers, and one-line delivery history.
4. Verify work in proportion to risk. Stage explicit owned paths only; inspect staged diff; commit and push only when the user has authorized it. Never use `git add -A` in a shared tree.
5. State what remains open and the exact next action.

## Ledger Rules

- Put temporary proposals, review requests, and cross-agent questions in **Review Ledger**, not `AGENTS.md`.
- The agent that created a row owns its deletion after both agents have reviewed it or it is cleanly resolved.
- Move permanent, user-approved constraints to **User Decisions**. Do not delete those without user direction.
- If another agent's changes are unclear, leave the row open and ask rather than guessing.

From shabanakhtar/ai-agent-skills, MIT.

Signals

Stars
0
Forks
0
Open issues
0
Last commit (this folder)
Jul 11, 2026 (3 months ago)

From GitHub (fetched Oct 24, 2026).

Details

License
MIT
First seen
Oct 24, 2026
Last checked
Oct 24, 2026

More from shabanakhtar

Done for you

Rather not set it up yourself?

Chalk Labs scopes, builds and runs agents for your team: the task audit, the eval harness, the integrations and the weekly ops report. Fixed scope, priced before we start.

See how we build AI agents →Read the case studies

Chalk Labs is a two-founder AI and growth studio. We build AI agents, CRMs and MVPs for Web3 and AI companies, and have shipped 50+ product and token launches.

launches shipped
50+
launches shipped
media placements
500+
media placements
founders on every account
2
founders on every account

Press we've placed: Forbes · CoinDesk · Decrypt · Cointelegraph · The Block

Rahil Jain, Co-founder, Chalk Labs. Ex-Goldman Sachs engineer. You talk to the person who builds it.

Get agent-sync running in your business

Tell us what you want to automate. A founder replies within one working day with a plan, not a pitch deck.

Goes to the Chalk Labs founders, not the newsletter. Privacy

Need help deploying agent-sync?

CallGet help
Cite this page

agent-sync skill. Agent Mart. https://agentmart.dev/skills/shabanakhtar-agent-sync (last updated 2026-10-24)