Runtime health
Manages runtime health, deployment, incidents, rollback, and monitoring. Use during deployment verification and incident triage.From its SKILL.md
npx -y skills add fworks-tech/agenthood --skill runtime-healthAssembled from the repository path, not quoted from the project. Check it against their README if it does not work.
2 things to look at
- 2 stars2 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.
- runs commandsInstructs the agent to run 5 commands, including `agenthood status` and 4 more.
What its file declares
Copied from the file, not written here
The file declares its own license as MIT. That is the author’s claim about this one file, and it is not the same thing as the license GitHub reports for the repository, which is listed with the other numbers below.
SKILL.md
3.4 KB, 716 tokens by cl100k_base, as published. Nobody here has run it
The Operator
Overview
The Operator watches over every running instance. When a deployment fails, a health check degrades, or a rollback is needed, the Operator is the first responder. It does not debug — it triages. It does not plan — it executes. The Operator keeps the runtime healthy by running diagnostics, performing rollbacks, and escalating to The Debugger when root cause is needed. Health is not a goal; it is a practice. The Operator makes practice routine.
When to Use
- When a deployment needs verification or rollback
- When runtime health checks are failing
- When monitoring metrics indicate degraded performance
- After running
agenthood verifyto lock member state - Before and after
agenthood rollbackto validate the result - When a member SKILL.md fails verification
- When a session needs runtime health diagnostics
Process
1. Assess Health
Run agenthood status to gather current state:
- Member count against registry
- Decision log and checkpoint counts
- Lockfile presence and validity
- Memory store initialization
2. Verify Integrity
If lockfile exists, verify current state matches locked state:
- Use
agenthood verifyto check member SKILL.md files - If verification passes, no action needed
- If verification fails, identify which members drifted
3. Initiate Rollback
When drift is detected:
- Run
agenthood rollback --dry-runto preview what would change - Run
agenthood rollbackto restore locked state - Run
agenthood verifyto confirm restoration
4. Escalate
If rollback fails or the issue is not member-related:
- Document the failure in the decision log
- Escalate to The Debugger for root cause analysis
- Notify The Herald if a release adjustment is needed
5. Document
Record the operation outcome:
- What was detected
- What was done (verify, rollback, status)
- Whether escalation was needed
Red Flags
- Verification passes but runtime still fails — the problem is not member drift
- Rollback restores files but
verifystill fails — lockfile is stale - A health check fails immediately after deployment before a lockfile was generated — no baseline exists
- Multiple members drift simultaneously — suggests a systemic issue, not a per-member one
- Rollback reverts unrelated files — git history is dirty (uncommitted changes)
Rationalizations
| What you think | What The Operator knows |
|---|---|
| "I can just git revert the change" | git revert rewrites history. Rollback preserves the lockfile as the source of truth and only touches member files. |
| "The tests pass, so everything is fine" | Tests verify correctness. The lockfile verifies integrity. They are orthogonal. |
| "I don't need to lock after deployment" | Without a lockfile, rollback has no target. Lock every deployment. |
| "One member failing is a minor issue" | Drift is contagious — one corrupted member degrades the Society's consensus. Roll back early. |
Verification
The operation is complete when:
-
agenthood statusshows all expected members present -
agenthood verifypasses for all members - Lockfile exists and matches current state
- Decision log records the operation
- Escalation path is clear (Debugger, Herald) if the issue recurred
What ships with it
Read from the repository
Just SKILL.md. No reference files, no scripts.
Gives 0 of the 12 instructions most ship operate skills give in 716 tokens
Counted across 1,077 of the 1,713 authors here whose files we hold, read 2026-09-06
- Create GitHub releasein 44 of 1077, across 43 files
- Run the test suitein 30 of 1077, across 25 files
- Create and push git tagin 27 of 1077, across 26 files
- Push commits and tagsin 27 of 1077
- Create annotated tagin 25 of 1077, across 22 files
- Ensure working tree is cleanin 24 of 1077
- Check for product marketing context firstin 23 of 1077, across 6 files
- Commit version bump changesin 22 of 1077, across 21 files
- Update CHANGELOG.mdin 21 of 1077, across 20 files
- Structure launch marketing across three channel typesin 20 of 1077, across 5 files
- Commit and tag the releasein 20 of 1077, across 18 files
- Update the CHANGELOG for new releasesin 19 of 1077
Said here and by no other author read
- Run agenthood status to gather current state
- Verify current state matches locked state
- Preview changes using agenthood rollback dry run
- Run agenthood rollback to restore locked state
- Confirm restoration using agenthood verify
- Document the failure in the decision log
Grouped from the skills themselves: near-identical wordings counted once, and counted by distinct author, so one author publishing three of these counts once. Length counted with cl100k_base; the agent that loads this file may tokenize it differently.