Report Memory plugin operational health via read-only JSON healthcheck diagnostics.
npx skills add https://github.com/reddb-io/red-skills --skill health-reddb-io
Please help me install this Agent Skill. Skill: health Source: https://github.com/reddb-io/red-skills/tree/main/plugins/memory/skills/core/health Command: npx skills add https://github.com/reddb-io/red-skills --skill health-reddb-io
bootstrap.mjs health --json and interprets every counter against the init-chain state taxonomy without mutating any file. - Operational counters: Reports initialization status, graph mode, graph freshness, Skill telemetry rollups, ranked proposal candidates, high-priority proposals, and pending proposal files. - Actionable next steps: Surfaces recommendedNextActions so agents and CI pipelines follow concrete guidance instead of inventing recovery steps. - Use Case: In an automated workflow, run this healthcheck before improve skills --write-proposal to confirm memory is ready and to detect pending proposals that should be reviewed first. ## Quick Start Ask the agent to run the Memory plugin healthcheck in JSON mode and report the state, counters, and recommended next actions for the current repository.High-intent search queries and answers about installing and using this skill.
Run the healthcheck with `node bootstrap.mjs health --json` and inspect the `state` field and `recommendedNextActions`. The JSON output reports initialization, graph mode, freshness, and pending proposals without mutating anything.
The healthcheck reports initialized status, graphMode, skillTelemetry, graphFreshness, rollups, proposalCandidates, highPriorityProposals, pendingProposalFiles, topProposals, and recommendedNextActions. Match the state field against the init-chain state taxonomy.
Yes, pass `--root <dir>` to the health command to check a repository other than the current working directory. This is useful for CI jobs or multi-repository agent workflows.
No, the healthcheck is strictly read-only and never mutates graph, proposal, or Skill files. It is a diagnostic intended to run safely before improvement workflows or large agent runs.
Treat `attention` as actionable, not as failure. Follow the `recommendedNextActions` field, and when `pendingProposalFiles` is greater than zero, inspect them with `improve proposals list/show --json`.
NominexHQReport PMM memory repository health with git logs, config data, and file stats.
damacusAnalyze coding session transcripts to generate improvement proposals for skills and workflows.
vessuxGenerate read-only per-bead proposals for open UNREFINED inbox items.
agno-agiInspect host operating system, Python runtime, and local directory contents as JSON.