agentleFS
narenaryannarenaryan/agent-skills/skills/security/rbvm-guarded-remediation/SKILL.md

rbvm-guarded-remediation

Skill0 starsChanged 41 days agoMITView on GitHub

Use when a vulnerability finding is about to trigger an automated patch, configuration change, isolation action, or retry against live assets, especially when evidence may be stale or ownership is unclear.

The file

We split this skill into the rules it gives an agent

3 rules. About 0 are common, 2 are distinctive, and 1 is 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.

  • 1

    Very specific

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

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    Use the provider's documented idempotency mechanism.

    line 24
  2. Distinctive:

    Run the bounded canary; compare against an unaffected control and absolute health limits.

    line 25
  3. Very specific:

    Stop expansion on failure.

    line 26

≈ “Common” is an estimate: at least 1 in 5 of the 364,980 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

Guarded Remediation

A finding is not mutation approval. Remediation is a conditional transition against a particular resource generation; the triggering snapshot may be obsolete.

Execution gates

GateRequired decision
OwnershipIdentify accountable service owner, change authority, executor, exact asset scope, and approved action
Current stateRe-read resource identity/generation and vulnerable condition immediately before mutation; changed state requires replanning
Retry identityPersist an operation ID with target, desired state, and parameters; reuse it only for the same intent
ReversibilityRecord known-good artifact/configuration and tested recovery path; irreversible changes need explicit review
Blast radiusChoose representative canary, observation interval, health/SLO thresholds, and stop conditions before execution

Recipe

  1. Record finding evidence, observation time, target identity, owner, approved parameters, and rollback reference. Missing ownership or approval blocks execution.
  2. Revalidate live state. Already fixed means no-op; changed identity/parameters means stop. Use provider-conditional mutation or equivalent atomic preconditions against external writers; workflow leases protect only against cooperating writers.
  3. Use the provider's documented idempotency mechanism. Store request parameters with the key; reject mismatches. For an uncertain timeout, reconcile remote state before retrying an action without safe replay semantics.
  4. Run the bounded canary; compare against an unaffected control and absolute health limits. Missing traffic or telemetry means inconclusive, not pass. Expand only after sufficient representative observations.
  5. Stop expansion on failure; perform only authorized recovery, then revalidate vulnerability state. Pass deployment evidence to rbvm-verified-closure.

Example: a stale finding requests disabling public access, but the live resource is already private. Record a no-op; do not restart its workload merely because the queue delivered the event twice.

Sources: AWS Config auto-remediation documentation; AWS Builders' Library, Malcolm Featonby; Google SRE Workbook, chapter 16 (2018). Reviewed 2026-10-10; source mapping and scenarios: SOURCES.md.

Pitfalls

  • Reusing a key after changing the patch target conflates distinct intent.
  • Updating only live infrastructure lets its declarative controller undo the fix.
  • Rolling back to a vulnerable release restores availability while reopening security risk.
  • Treating emergency priority as permission to skip identity, approval, or recovery checks.

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.