indigoai-us/hq-core/hq-cowork-sync

Claude Code and agent skill by indigoai-us/hq-core

Trust tier 3
  • community
Source check pending

Checking that the SKILL.md file is still in the repository.

Run bidirectional HQ Sync from Cowork.

Install

npx vskill@latest install indigoai-us/hq-core/hq-cowork-sync

SKILL.md

/hq-cowork-sync — Sync HQ from a sandboxed agent

Runs the same sync engine the HQ Desktop App uses, but from inside a sandboxed Claude Code plugin host (Cowork, in particular). Walks every cloud-backed company in your local HQ, syncs in both directions against the vault, and writes conflict mirror files when divergence is detected.

Args: $ARGUMENTS — optional flags. Accepted shapes (parse free-text):

  • --company <slug> → scope to one company
  • --personal → sync only the caller's personal vault
  • --on-conflict overwrite|keep|abort → conflict strategy (default keep)
  • --message "<msg>" → optional journal message attached to push leg

Without args, syncs every membership + personal vault with --on-conflict keep.

When to use this instead of /hq-sync

  • You're in Cowork or another sandboxed plugin host. The sandbox can't see

~/.hq/cognito-tokens.json and hq isn't on PATH inside it, so the regular /hq-sync skill (which spawns npx ... hq-sync-runner) cannot run.

  • You want native MCP tool-call ergonomics — observable in the host's tool

log, not just shell output.

In a regular Claude Code session running on the host directly, prefer the unprefixed /hq-sync skill — same engine, fewer hops.

What you do

Step 1 — Confirm the host MCP is wired

Before the first sync of a session, call mcp__hq__hq_whoami. If it returns your identity + session expiry, the host MCP is live and authenticated. If it errors with "not signed in" or "tokens expired", tell the user to run hq login on the host machine (the sandboxed agent cannot do this itself — hq login opens a browser on the host).

Step 2 — Call the sync tool

Parse $ARGUMENTS into the hq_sync tool's input schema:

Arg patternTool input
--company <slug>{ "company": "<slug>" }
--personal{ "personal": true }
(neither){} — server defaults to --all
--on-conflict <strategy>{ "onConflict": "<strategy>" }
--message "<msg>"{ "message": "<msg>" }

Call mcp__hq__hq_sync with the assembled input. Sync output can be chunky (per-company push + pull legs) — surface a terse summary, not the full log, unless the user asked for verbose detail.

Step 3 — Report

Quiet by default. Surface:

  • Companies synced (count + names)
  • Conflicts written (count + the path to <hqRoot>/.hq-conflicts/index.json

if any — the user can run /resolve-conflicts next)

  • Errors from any individual company leg (don't swallow them)

If conflicts were written, mention /resolve-conflicts as the next step.

Why this skill exists

The unprefixed /hq-sync skill resolves HQ root via four config tiers and then execs the cloud-runner binary directly with the host's Cognito session. None of that works from Cowork's sandbox, where ~/.hq is not mounted and the host's Node binary isn't reachable. The hq-pack-cowork MCP server fixes that by running on the host (with full auth + binaries) and exposing hq_sync as a tool call the sandboxed agent can make. This skill is the thin in-session adapter for that tool.

Security

  • Security scanPassed, 90 of 100Pattern checks on the skill's files.
  • Model reviewPassed

Quality evaluation

  • Task evaluationNot runNo evaluation yet. Run one in Skill Studio.

Source

Popularity

Repository stars
87
Repository forks
17
7-day trend score
0

Stars and forks are for the whole GitHub repository, not this skill alone. Not refreshed yet.

Works with

Any agent vskill installs to: 56 targets in all. These 39 have a page on this site.

  • Claude Code
  • Cursor
  • GitHub Copilot
  • Windsurf
  • Codex
  • Gemini CLI
  • Cline
  • Amp
  • Roo Code
  • Goose
  • Continue
  • Kilo Code
  • Junie
  • Trae
  • OpenHands
  • Qwen Code
  • AI Agent
  • Tabnine Chat
  • Sourcegraph Cody
  • Amazon Q
  • Firebase Genkit
  • Sweep
  • Devin
  • Phind
  • Mentat
  • Claude Desktop
  • Aider
  • GPT Pilot
  • Plandex
  • bolt.new
  • v0
  • Replit Agent
  • Anse
  • TypingMind
  • BetterChatGPT
  • LibreChat
  • Warp AI
  • Pieces
  • Fig AI
Embed badge
vskill VERIFIED badge
[![vskill VERIFIED](https://verified-skill.com/api/v1/skills/indigoai-us/hq-core/hq-cowork-sync/badge)](https://verified-skill.com/skills/indigoai-us/hq-core/hq-cowork-sync)

Something wrong with this skill or its page? Tell us.