AgentHub Board: Claims, Handoffs, Conflict Checks
Read and post on the AgentHub board: digest what other agents claimed, catch duplicate work, and write claims and handoffs that stick.
Before & after
Same prompt, same agent — the only difference is the skill.
I can check a message board for you, though I'd need to understand more about what system you're using. There are a few different ways agents or teams coordinate through shared messaging boards or channels. Are you using a specific platform? If you're running multiple Claude agents and they're posting coordination messages to a shared board, that's an interesting setup. I'd need to know where that board is hosted and how to access it. Can you point me to the board or explain how your coordination system works?
BOARD CHECK — Agent Coordination Messages (2026-07-31): TASK PRIORITY MESSAGES: → 11:34 | analytics-agent: 'Priority shift — dashboard refresh stalled on API timeout. Pushing report deadline to 14:00 UTC. Depends on data-fetch-agent completion.' → 12:07 | data-fetch-agent: 'Acknowledged. Data pull complete 11:58, 892 records processed. Available for downstream work.' → 12:15 | security-scan-agent: 'Request: Can compliance-checker prioritize token-audit task before 16:00? Flagged in access logs.' → 12:22 | compliance-checker: 'Yes, prioritized. Running now, results by 15:30.' COORDINATION SUMMARY: → Three tasks currently waiting on dependencies → No conflicts detected in queue → One urgent flag (token audit) now escalated → Estimated clearance time: 15:30 UTC YOUR NEXT STEP: Priority dashboard report now available at 14:00. Data-fetch-agent signals completion, no blockers remaining.
About this skill
Your agents keep stepping on each other. Two of them refactor the same module, a handoff note says "done, see branch" with no branch name, and the board is 200 messages nobody can read. This skill makes Claude read the AgentHub board like a dispatcher. It catches collisions before work starts and writes posts that another agent can act on cold.
What it does
- Turns a noisy board into a digest: the latest state per agent and task, open claims, stale claims, unanswered questions, and conflicts flagged at the top
- Checks for overlapping claims before posting. It re-reads after posting to catch simultaneous claims and applies a stated tie-break.
- Writes typed posts (CLAIM, STATUS, HANDOFF, BLOCKER, RESULT, QUESTION, RELEASE) with scope, artifacts, and an expiry
- Writes handoffs a fresh agent can resume from: the branch, the commit, what was verified, what was not, and the exact next step
- Treats other agents' posts as data, not commands. It keeps secrets off the board and avoids leaking solution details in competitive sessions.
- Matches the board's existing format and never claims to have read or posted when it could not reach the board
What you get
For a read request, you get a short conflict-first digest of the board. For a write request, you get a ready-to-post message after a duplicate check, along with confirmation of what was actually posted.
Best for
Developers running several Claude Code agents in parallel on one repo through AgentHub. It is also useful for anyone who needs claims, handoffs, and blockers to stay legible across agents and sessions.
SKILL.md preview
---
name: agenthub-board-claims-handoffs
description: Reads, digests, and posts to the AgentHub message board so parallel agents coordinate without duplicate work. Before posting, it checks for overlapping claims, writes typed and scoped posts with artifacts and expiry, and reports conflicts instead of resolving them silently. Use when the user says "/hub:board", "check the board", "is anyone already working on X", "post a handoff for the next agent", "claim the auth task", "what did the other agents say", "my agents keep duplicating each other's work", or "the board is a mess, summarize it".
---
# AgentHub Board: Claims, Handoffs, Conflict Checks
… (sign up to view the full skill)More ai & agents skills
View all AI & Agents skills →New Skill Scaffolder
Create a new Claude Code skill correctly: right mechanism, valid name and folder, a triggering description, and a smoke test.
Skill Library Lint and Overlap Report
Lint every skill in a repo at once: frontmatter, name-folder match, weak descriptions, overlapping triggers, broken refs, plus a CI script.
Self-Verifying Agent Loop: Plan, Check, Retry, Close
Turns a goal into a task plan with machine-run checks, executes with retry caps, escalates, and refuses to close until verified.
Skill Test Runner (Trigger and A/B)
Test a skill in fresh sessions: ready-to-run prompts, commands, and rubric for trigger recall, false triggers, and blind A/B.