Integrated Resource Plan Drafter

• Outcome: Pass
• Latest audit: Sep 16, 2026 · 9:04 AM UTC
• Version: 0.1.1

Overview

This is a drafting-only utility-regulatory skill that is clearly scoped to producing unfiled drafts for human verification and signature.

Appropriate for drafting support, but users should still keep protected utility workpapers, customer-identifying data, contract prices, and confidential rate-impact details within authorized systems and have regulatory counsel/resource-planning staff verify all filing content before use.

A.I.G

Vulnerability Patterns

• Skill Instruction Hijacking: Alters the agent's session goals or safety constraints when the skill loads
• Agent Memory Poisoning: Writes attacker-controlled rules into memory that affect later sessions
• Remote Payload Retrieval and Execution: Fetches external code whose behavior can change after review
• Embedded Malicious Code: Ships malicious scripts inside the skill and executes them locally
• Unauthorized Access and Privilege Escalation: Obtains permissions beyond the task's legitimate needs

SkillSpector

Vulnerability Patterns

• Excessive Agency: Unrestricted Tool Access, Autonomous Decision Making, Scope Creep
• Prompt Injection: Instruction Override, Hidden Instructions, Exfiltration Commands
• Data Exfiltration: External Transmission, Env Variable Harvesting, File System Enumeration
• Privilege Escalation: Excessive Permissions, Sudo/Root Execution, Credential Access
• Supply Chain: Unpinned Dependencies, External Script Fetching, Obfuscated Code

Findings (1)

Autonomous Decision Making

Medium

• Category: Excessive Agency
• Confidence: 75% confidence

Finding

Skill enables autonomous high-impact decisions without human-in-the-loop verification. Critical operations (destructive commands, financial transactions, data deletion) should require explicit user confirmation.

Content

Scanner excerpt · SKILL.md (reported line 340)May include surrounding context.

md

- **DRAFT only.** Every chapter, appendix, and the cover letter must be labeled `DRAFT — for filing utility regulatory team to verify and sign`. The skill produces no served filing.
- **The filing utility signs, not the skill.** Even if the user is a regulatory officer, the signature block remains unsigned in the DRAFT. Service is performed by the filing utility under its own filing protocol.
- **Never opine that a portfolio is "least-cost / best-fit".** That determination is the filing utility's and ultimately the PUC's. The skill reports cost, reliability, and policy-compliance metrics and lets the filing utility frame the portfolio characterization.
- **Never affirm model output without verification.** Every capacity-expansion model output, ELCC value, transmission-study output, and load-forecast trajectory must be traceable to a workpaper, model run, or third-party study, and is the filing utility's resource-planning team's responsibility to verify.
- **Never assume the RA program rules.** Confirm the regional RA program (WRAP, CAISO, MISO, PJM, SPP, ISO-NE, NYISO, ERCOT) and the capacity-counting rules with the filing utility. Do not infer them from a peer utility's filing.
- **Never blend public and confidential text.** Every chapter, table, and appendix is marked in the confidentiality-treatment table. The public-redacted version uses the controlling protective order's placeholder language, not the underlying figure.
- **Never silently true-up load reconciliations.** Where gross load → load modifiers → managed load → LSE-assigned load does not close, log the gap as an open item.