agentleFS
ghivelyghively/infra-ops/skills/instinct-rollback/SKILL.md

instinct-rollback

Skill0 starsChanged 5 months agoMITView on GitHub

Roll back a governed instinct to a previous version or deactivate it, with two-person approval for compliance/HSA instincts and a full governance-ledger audit trail. Triggers on: instinct rollback, deactivate instinct, revert instinct, rollback pattern, /instinct-rollback.

The file

We split this skill into the rules it gives an agent

2 rules. About 0 are common, 2 are distinctive, and 0 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈2

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 0

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    Use this skill when a promoted instinct turns out to be incorrect, harmful, or no longer applicable and must be reverted to a prior version …

    line 15
  2. Distinctive:

    Every rollback/deactivation is recorded in the governance ledger.

    line 64

≈ “Common” is an estimate: at least 1 in 5 of the 350,495 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

Instinct Rollback Skill

When to Use

Use this skill when a promoted instinct turns out to be incorrect, harmful, or no longer applicable and must be reverted to a prior version or deactivated. Reach for it on prompts like "roll back that instinct", "deactivate the instinct", or /instinct-rollback. Deactivation takes effect for the next session — it does not retroactively undo work already produced under the instinct.

How It Works

Rollback is governed and audited; the agent never hand-edits ledger files. The scripts/lib/instinct-ledger.js CLI updates the entry (status plus a rollback: or deactivated: block) and logs an instinct-rollback event to the unified State Store.

Approval rules:

  • Routine instinct — at least one --approver.
  • Compliance-related or HSA (hsa/in-zone) instinct — two distinct approvers (dual control). The CLI rejects a single-approver rollback for these.
  • Reactivation of a deactivated compliance instinct is at least as strict as its rollback (re-promote through /instinct-promote with the same gate; HSA needs dual control again).

Zones are corporate / hsa (legacy corpor / in-zone accepted).

Examples

List, then roll a corporate instinct back a version

node "${CLAUDE_PLUGIN_ROOT}/scripts/lib/instinct-ledger.js" --list --zone corporate
node "${CLAUDE_PLUGIN_ROOT}/scripts/lib/instinct-ledger.js" --rollback \
  --id fqcn-001 --zone corporate --version 2 \
  --reason "Caused false positives on legacy roles" --approvers senior-op-1

Deactivate a compliance/HSA instinct (dual control enforced)

node "${CLAUDE_PLUGIN_ROOT}/scripts/lib/instinct-ledger.js" --rollback \
  --id perso-x --zone hsa --deactivate \
  --reason "Pattern conflicts with operator-only requirement (PCI CP §X)" \
  --approvers senior-op-1,cpsa-assessor
# A single approver here is rejected: compliance/HSA rollback needs two distinct approvers.

Trust boundary

  • The agent proposes the rollback; humans approve it. Compliance/HSA items are never rolled back or reactivated on a single approval.
  • Reasons must be specific; for compliance items state the compliance impact.
  • Every rollback/deactivation is recorded in the governance ledger.

Deep Reference

When to Rollback vs. Supersede

  • Rollback (deactivate): the instinct was wrong or is no longer valid
  • Supersede (create new version): the instinct was right but has been refined

For a supersede: create a new instinct with updated content; in the new instinct's evidence field, reference the original instinct ID and explain the refinement. Then rollback the original.

Rollback Governance Requirements

Rollbacks are governance events and require:

  1. A human approver (the rollback initiator does not self-approve)
  2. A reason string explaining why the instinct is being deactivated
  3. Logging to the governance ledger (automatic via instinct-ledger.js --rollback)

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.