agentsclimarketplace
Sign in

Watercooler health

Skill mostlyharmless-ai/watercooler/skills/watercooler-health

Check watercooler system health — MCP server, baseline graph (T1), git auth, GitHub rate limit, and daemons. Use when syncs break or anything in the watercooler stack behaves unexpectedly.From its SKILL.md

Install
npx -y skills add mostlyharmless-ai/watercooler --skill watercooler-health

Assembled from the repository path, not quoted from the project. Check it against their README if it does not work.

One thing to look at

  • 13 stars13 stars. Stars are a popularity signal and not a quality one, but at this level it is likely that nobody has read this closely except its author, and you would be relying on your own review.

SKILL.md

3.1 KB, as published. Nobody here has run it

Watercooler Health Check

Run a full health check across all watercooler subsystems.

Steps

  1. Load diagnostic tools in parallel:

    ToolSearch: select:mcp__watercooler__watercooler_health
ToolSearch: select:mcp__watercooler__watercooler_baseline_graph

  2. Run checks in parallel:

    mcp__watercooler__watercooler_health(code_path="<repo root>")
mcp__watercooler__watercooler_health(code_path="<repo root>", detail="identity")
mcp__watercooler__watercooler_baseline_graph(scope="sync", code_path="<repo root>")

    The detail="identity" call returns the resolved agent identity and a write-readiness assessment (threads-dir writability + agent_func guidance) — this is the folded-in former whoami check.

  3. Report status — organize output into these sections:

    Server Health (from watercooler_health):

    • Server version and status
    • Agent identity and threads directory
    • Code branch and auto-branch mode
    • Python / fastmcp versions
    • Write-readiness (from watercooler_health(detail="identity")): resolved identity and whether the threads dir is writable

    Graph Services (from watercooler_health):

    • Summaries enabled, LLM service URL and availability
    • Embeddings enabled, embedding service URL and availability

    Backend Services (from watercooler_health):

    • Auto-start services (llm, embedding, falkordb) with state icons
    • Include startup time (ms) for running services
    • Surface setup instructions for any failed services

    Daemons (from watercooler_health):

    • Each daemon: state, interval, ticks, findings, errors

    Thread Storage (from watercooler_health):

    • Mode (orphan worktree), path, code branch

    Git Authentication (from watercooler_health):

    • Protocol (ssh/https), connectivity status
    • Credential helper or SSH agent/key status
    • Any auth warnings and recommended fixes

    GitHub (from watercooler_health):

    • gh CLI version (warn if outdated)
    • API rate limit: remaining/total (%) and reset time
    • Any warnings and recommendations

    T1 — Baseline Graph (from baseline_graph, scope="sync"):

    • Total / synced / stale / error threads
    • Recommendations (if any)

  4. Suggest fixes for common issues:

    T1 issues:

    • Stale threads: run watercooler_baseline_graph (scope="sync") and reconcile stale entries

    Git / GitHub issues:

    • SSH without agent: eval "$(ssh-agent -s)" && ssh-add
    • Expired GitHub CLI token: gh auth login -h github.com --web
    • Rate limited: pause automated operations; check reset time

Example Invocations

  • /watercooler-health — Full system health check
  • Use when syncs or push/pull operations fail
  • Use when daemons are not running or producing no findings

Keep looking

Skills are one crate of 627,801. Ordering is by how many stacks a row turns up in, so the top of any crate is what has actually been picked rather than what has the most stars.