Energy AI agent · NERC CIP evidence
NERC CIP Compliance Evidence AI Agent
Carry the evidence each CIP interval consumes — the patch date, the evaluation, the baseline, the quarterly access check — and hold it for the CIP Senior Manager named under CIP-003-9 R3.
- Typical delivery
- 4–6 weeks
- Deployment
- Your stack
- Named manager
- Interval-bound
- After launch
- Agent Care
What this agent does
Assembles the evidence, not the determination
In
- 01
A patch publishes, and CIP-007-6 Part 2.2 allows 35 calendar days to evaluate it for applicability.
- 02
An evaluation completes, and Part 2.3 gives 35 more days to apply or to date a mitigation plan.
Reason
- 03
A mitigation plan runs long, and only the CIP Senior Manager or delegate may extend it under Part 2.4.
- 04
A change completes, and CIP-010-4 Part 1.3 allows 30 calendar days to update the baseline.
- 05
An access review falls due, and CIP-004-7 Part 4.2 wants verification each calendar quarter.
Decide
- 06
A termination action lands, and CIP-004-7 Part 5.1 leaves 24 hours to revoke the access.
- 07
An incident is judged, and the CIP-008-6 Part 4.2 hour runs from determination, not detection.
Out
- 08
A manager changes, and CIP-003-9 R3 wants the new name documented within 30 calendar days.
- 09
Execute write actions only inside the approval boundaries agreed during implementation.
Product statement
The agent assembles and tracks intervals. The named person is the CIP Senior Manager, who approves — the self-certification itself carries no attestation.
Example workflow
One interval, evidence to approval
- 1Interval evidence receivedPatch bulletins, baselines, access records or termination feeds
- 2Interval context assembledThe system, its impact rating, the standard part it answers to and the date the clock started
- 3Evidence set draftedThe interval, its records, the gaps and completeness
- 4Controls appliedInterval checks, evidence-sufficiency checks, date checks and completeness confidence
No human action required
Stages 1 to 4 run unaided, and nothing is approved at any of them — the agent is assembling, and the compliance lane opens at the completeness gate.
Goes to the CIP Senior Manager to approve.
Adds a compliance analyst read first.
Compliance review
The set is held with its interval, its gaps and the systems they sit on.
Approve · Append evidence · Send to compliance
Approved — by the CIP Senior Manager
- 6Asset and access records updatedOnly where write access and records policy allow it
- 7Outcome evaluatedEvidence completeness, interval coverage, analyst corrections and what review found
Corrections
Each analyst correction is counted in the evaluation.
What should not run autonomously
Human approval stays in control
Outside the boundary — human approval required
8 items
- Determining that an incident is reportable.
- Approving cyber security policies under CIP-003-9 R1.
- Signing a self-certification or any CMEP submission.
- Extending a mitigation plan under CIP-007-6 Part 2.4.
Automation boundary
Agent acts unaided
- ✓Hold the record each CIP interval consumes, with the date it carries.
- ✓Track each interval against the standard that sets it.
- ✓Chase the missing record before the calendar-day window closes.
- ✓Flag the determination criteria that start the one-hour.
Nothing is approved or filed except by a named person, inside the agreed boundaries.
- Judging whether an interval was in fact met.
- Telling a Regional Entity the estate is compliant.
- Setting the impact rating a system is held at.
- Changes to access, baselines or firewall rules.
Example output
One interval, annotated
Our control room copilot leaves the command to the operator; this record is what one CIP interval consumed.
Evidence set · single interval
Illustrative example- Interval
- Patch evaluation, Part 2.2
- Recorded as
- Evaluated for applicability, not applicable
- System
- Medium impact
- Evidence of record
- Vendor bulletin, 3 August 2026
- Confidence
- Held unapproved
- Held for
- The CIP Senior Manager, by name
As received
Taken from the vendor bulletin and the patch record — it reaches as far as those sources do.
What the record holds
- Vendor bulletin date
- Evaluation record
- Mitigation plan, dated
Why no interval call here
Whether the window was met is a CIP-007-6 judgement, not a model output.
What the score decides
Below the configured threshold the set picks up a compliance read before the manager sees it.
Action
ApproveAppend evidenceSend to complianceValue
Where AI adds value
The same four claims, placed at the point in the workflow where each one applies.
Where the value lands
Value 01 – 04
Every interval
From the system that owes it
Where the evidence is used
Our compliance reporting assistant works a register and names no regime; this page is one regime, with its own clocks and its own named role.
The clock starts when you say so
CIP-008-6 Part 4.2 runs one hour from determination under your own R1 Part 1.2 criteria, not from detection.
What was checked, and not found
NERC publishes neither CIP audit reports nor noncompliance filings — they are treated as critical energy infrastructure information — and no aggregate total is published.
Build an evidence trail
The evidence, the interval it belongs to and the manager who approved stay on the record.
Integrations
Typical integrations
Five system groups connect to the same agent. Which of them are in scope is decided in discovery.
Patch and vulnerability sources
- Vendor bulletins
- CVE feeds
- Patch management systems
Configuration and baselines
- CMDB
- configuration monitoring
- Change records
Access and personnel
- Identity provider
- HR feed
- Termination and transfer records
Agent
NERC CIP compliance evidence
- Reads the intervals
- Assembles the evidence
- Holds for the manager
Incident and case systems
- SIEM
- ticketing
- Determination criteria records
Observability & evaluation
- OpenTelemetry
- Langfuse
- Supported monitoring/evaluation sources
Integration availability depends on the client's existing systems and API access.
Agent controls
Six meshes between the model and the manager
Six meshes, coarsest first of all. Whatever drops through every one is set out in the map below.
L6 · Outermost — last line of defence
Inward → L1 · closest to the model
- L1 – L2
- Test whether a set may stand
- L3
- Puts the approval in a person's hands
- L4 – L5
- Keep the interval and the evidence behind it
- L6
- Holds the determination unmade when signals degrade
How Nestack evaluates it
Evaluate the whole assembly — not only the evidence set that comes out.
Coverage runs the whole depth of the workflow, and every layer is cut by slice.
Surface — the set an auditor reads
- E1Final-output evaluationDid the set record what each interval actually consumed?
- E2Step-level evaluationDid the agent read the right system, the right part and the live configuration?
- E3Tool evaluationDid it read and write the correct asset record and the correct interval?
- E4Confidence calibrationDo low-confidence sets actually attract more compliance corrections?
- E5Slice evaluationHow does performance change across specific impact ratings?
- E6Business outcomeHow many sets needed a correction before the manager approved?
Floor — the outcome the entity answers for
Failure modes
Where each failure originates in the agent
Seven failure modes, each set at the stage where it first appears.
Agent lifecycle
Direction of processing →
01 · Retrieval
1 mode
- IX-03
Stale interval read
The configuration read is not the one now in service.
Stage gathers
The systems, the intervals, the records and the dates
02 · Reasoning
2 modes
- IX-04
Interval asserted, not shown
An interval is called met without its evidence.
- IX-06
Future standard read as live
CIP-015-1 is worked as though enforceable now.
Stage proposes
The intervals, their records and completeness
03 · Tool / write
2 modes
- IX-02
Thin set passed forward
A set moves on without the compliance read.
- IX-05
Evidence bound to wrong asset
A record is filed against the wrong system.
Stage writes
Only where write access and approval policy allow it
04 · Output
1 mode
- IX-01
Approved, evidence unrecorded
The record shows approval but not what supported it.
Stage returns
The set a manager approves and an auditor reads
05 · Change / Version
1 mode
- IX-07
Silent interval regression
A configuration change moves the interval, not the record.
Stage tracks
Model, prompt, interval rules and evidence fields
- Sev-1 · an interval approved on no evidence
- Sev-2 · wrong evidence reaches the record
- Sev-3 · source degrades, set holds unapproved
Affected slices
Low impact absorbs the corrections
A rating-level interval-evidence figure can read clean while low impact systems carry most of the corrections. Nestack reports the correction rate by impact rating, not only in total.
Slice performance — reported separately, not only in aggregate
Illustrative example| Slice | Failure rate | Lift | Lift vs. threshold | Status |
|---|---|---|---|---|
| Low impact systems | 10.2% | 3.6× | Review | |
| Medium impact at Control Centers | 7.6% | 2.7× | Review | |
| Medium impact with routable access | 4.7% | 1.7× | Watch | |
| High impact Control Centers | 1.6% | 0.6× | Normal |
Bar: correction-rate lift vs. high-impact Control Center baseline · scale 0–4.0× · tick marks the 2.0× review threshold
2 of 4 slices over threshold
Evidence-linked improvement
What a missed window costs
A cycle closes when the missed evaluation window is a regression case. That suite is what the next interval worked is measured against.
Improvement cycle · five stages
Switchback — the path turns at Improve and returns at Learn
- 01
Detect
Correction rate rises on low impact systems.
- 02
Diagnose
The patch that published on a Friday and was evaluated thirty-six days later is read back until one cause remains.
- 03
Improve
The change ships numbered, and the intervals that forced it ride with it.
- 04
Verify
Nothing releases while one touched system case is still red.
- 05
Learn
It is retained for good, and the interval rules are amended in that same commit.
Learn → Detect
The return edge. The next interval is measured against a suite one case longer.
Typical build scope
Twelve workstreams across six weeks
The build scope read against the delivery timeline. Week structure follows the six-week plan — discovery, sources, evidence assembly, evaluation, integration, then production validation and handover.
Workstream
Week 1Week 2Week 3Week 4Week 5Week 6
- 01CIP-interval discovery and automation-boundary work.Week 1
- 02Patch, baseline and access sources.Week 1, Week 2
- 03Interval-to-standard and evidence-coverage mapping.Week 1, Week 2
- 04Interval evidence ingestion.Week 2
- 05System, part and record binding.Week 2, Week 3
- 06Completeness scoring and review routing.Week 3
- 07Manager approval workflow.Week 3, Week 4
- 08Asset and access-system integration.Week 5
- 09Interval and evidence cases.Week 4, Week 5
- 10Guardrails and approval controls.Week 4
- 11Interval-trail instrumentation.Week 5, Week 6
- 12Deployment, documentation and Agent Care handover.Week 6
12 workstreams · 6 weeks · bar shows the weeks a workstream is active — several run in parallel
Final scope and sequence confirmed in discovery
Engagement tiers
What each tier includes
Rows are the capabilities named in each tier's scope. Higher tiers include everything below them.
| Capability✓ in scope · — not at this tier | PilotOne rating, one year | ProductionProduction compliance workflow | AdvancedMultiple ratings / entities |
|---|---|---|---|
| Introduced at Pilot | |||
| Evidence assembly to your intervals | ✓ | ✓ | ✓ |
| CIP Senior Manager release | ✓ | ✓ | ✓ |
| Asset-inventory baseline | ✓ | ✓ | ✓ |
| Introduced at Production | |||
| Reporting by impact rating | — | ✓ | ✓ |
| Approval workflow in your systems | — | ✓ | ✓ |
| Approved write-back | — | ✓ | ✓ |
| Configuration-system integration | — | ✓ | ✓ |
| Introduced at Advanced | |||
| Multi-entity registrations | — | — | ✓ |
| Cross-standard evidence packs | — | — | ✓ |
| Large asset estates | — | — | ✓ |
| Multi-standard interval controls | — | — | ✓ |
| Build price | From $5,000 | From $8,000 | Custom quote |
- Final build price
- Confirmed after discovery based on integrations, workflow complexity, transaction volume, approval controls and deployment requirements.
- Separate from build
- Build pricing is separate from recurring Agent Care, which covers managed monitoring, evaluations, incidents and verified improvements after launch.
What we need from you
What you bring, and what we build with it
Each input maps to a piece of build scope and a week in the delivery timeline.
You bring
We build with it
01Your registered systems and their impact ratings
→Asset-inventory mapping and interval capture
Week 102Representative patch, baseline and access records
→Record binding, interval logic and the evidence baseline
Week 203Your determination criteria under R1 Part 1.2
→Interval mapping, system binding and the automation boundary
Week 104Access to relevant APIs, feeds or exports
→Patch, configuration and access-source assessment, then integration setup
Week 205Intervals you would not want reconstructed
→Evaluation cases and failure-mode testing
Week 406What no evidence set may establish
→Completeness scoring, review routing, guardrails and release controls
Week 307A named CIP Senior Manager to approve
→Manager approval workflow, then pilot and production validation
Weeks 5–6
Nothing else is required
Deployment, documentation and Agent Care handover are ours.
Delivery timeline
Four phases across six weeks
These bands are the weeks each phase honestly costs, and the fifth one therefore has to carry two.
Phase
W1
W2
W3
W4
W5
W6
- DiscoveryW1
- BuildW2 – W3
- EvaluateW4 – W5
- Pilot & LaunchW5 – W6
Week focus
W1CIP workflow discovery, interval mapping and the automation boundary
W2Source integration and the interval-evidence baseline
W3Evidence assembly, interval logic and release controls
W4Evaluation suite, interval cases and failure-mode testing
W5Record integration, pilot intervals and targeted corrections
W6One compliance year run under the CIP senior manager, then Agent Care handover
- Reading the band
- Each bar covers only the weeks its own work is named for. The fifth week carries two because the work does.
- At the end of W6
- Validation closes on live intervals, and Agent Care picks up the watch.
- Duration
- Six-week plan shown · typical delivery 4–6 weeks depending on scope confirmed in discovery.
Next step · Energy AI agent
Build a CIP evidence agent around the intervals your estate already owes.
Show us one interval and the record it consumed. If patch evaluation and mitigation planning sit apart across your estate, then the Part 2.3 window is kept on trust. Your self-certification calendar is your Regional Entity's; plant OT monitoring is a different agent.
- Typical delivery
- 4–6 weeks
- Deployment
- Your asset inventory
- Named manager
- Interval-bound
- After launch
- Agent Care available
Nestack Agents · NERC CIP evidence
AGT-EN-10 · Agent Care available after launch