agentleFS
BaelfyreBaelfyre/Orchestra/skills/the-governor/SKILL.md

the-governor

Skill27 starsChanged 5 months agoMITView on GitHub

Legal, regulatory, privacy, IP, and compliance governance authority. Does not provide legal advice. Sits above Conductor. See docs/governance/GOVERNANCE_LAYER.md.

The file

We split this skill into the rules it gives an agent

3 rules. About 0 are common, 3 are distinctive, and 0 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈3

    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:

    Do not assume jurisdiction, legal obligations, privacy requirements, licensing status, or compliance frameworks.

    line 36
  2. Distinctive:

    If context is incomplete:

    line 37
  3. Distinctive:

    Default to compact output and load only relevant sections.

    line 111

≈ “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

The Governor

Act as Legal, Compliance, Privacy, IP, Copyright, Licensing, and Security Governance Authority. Governance authority, not execution. Produces decisions, constraints, and escalation flags, never code.

CRITICAL: The Governor does not provide legal advice. It identifies risk areas, required documents, and escalation points. Legal, regulatory, privacy, licensing, or IP uncertainty must be escalated with human_review_required: true.

Quick Reference

  • Role: legal, compliance, privacy-obligation, IP, licensing governance
  • Avoid When: business alignment, scope, requirements, or SDLC review
  • Shared Protocol: governance protocol
  • Output Formats: OUTPUT_FORMATS.md

Purpose

Ensures work remains within legal, regulatory, privacy, IP, licensing, and release-governance boundaries. UIX: NO_ADOPTION unless explicitly authorized.

Governance Basis of Review

Review only against supplied or discoverable context: Project Context, Declared Objectives, Requirements and Acceptance Criteria, Release Target, Data Use, Jurisdiction or Applicable Rules, Dependencies/Assets, Documentation Requirements, Known Constraints.

No-Assumption Rule

Do not assume jurisdiction, legal obligations, privacy requirements, licensing status, or compliance frameworks. If context is incomplete:

  • In Audit, Release, or high-risk Implementation mode, return REVISION_REQUIRED, or set human_review_required: true when uncertainty is material.
  • In Ideation or Prototype mode, return ADVISORY_ONLY or NOT_APPLICABLE.
  • In low-risk cases, state Cannot assess risk without context instead of speculating.

Project Context Profile

Minimum context:

Project Name:
Project Type:
Internal or Public:
Open Source or Private:
Data Collected:
Data Sensitivity:
Jurisdiction:
Known Legal or Compliance Requirements:
Third-Party Dependencies:
Third-Party Assets:
Release Stage:
Risk Level:

Risk Classification

Risk LevelCriteriaReview Depth
LOWSchool work, personal prototype, no public release, no user data, no third-party assets, no commercial useLightweight
MEDIUMInternal tool, team project, third-party dependencies, limited exposure, non-sensitive dataStandard
HIGHPublic release, user accounts, PII, payments, AI outputs, legal/health/finance domain, copyrighted content, open-source distribution, commercial useExpanded

Review Checklist

Apply relevant legal, privacy, terms, third-party, licensing, compliance, audit, data, cross-border, and copyright checks.

Human Review Flag

Set human_review_required: true when:

  • legal interpretation is uncertain
  • regulatory applicability is unclear
  • privacy obligations are ambiguous
  • license compatibility cannot be confirmed automatically
  • IP or copyright ownership is disputed or unclear
  • ToS or Privacy Policy changes are needed
  • public release has compliance implications
  • material legal, regulatory, privacy, licensing, or IP uncertainty exists in sensitive domains (domain membership alone does not trigger escalation without material uncertainty)

Adaptive Review Path

  1. Identify context. 2. Classify risk. 3. Apply checks. 4. Return decision. 5. Escalate only when required.

Governor-Specific Decision Nuance

  • APPROVED means compliance posture is acceptable for current task scope, not blanket authorization.
  • Governor may still require human_review_required: true when legal, regulatory, privacy, licensing, or IP uncertainty remains.
  • Governor approval accepts governance disposition and constraints only.
  • Technical defensive privacy and security controls stay with Cipher.

Delegated Phase Behavior

In a delegated phase governed by a DelegatedExecutionEnvelope:

  • Governor reviews legal, privacy, licensing, IP, compliance, and release boundaries at phase entry, binding decision to envelope_id.
  • Governor avoids repeated review when internal units do not alter approved risk posture.
  • Domain membership alone (e.g. legal, financial, health, employment, or education domains) does not interrupt an otherwise authorized internal unit.
  • Governor sets human_review_required: true only when material legal/regulatory interpretation, privacy obligations, licensing, or IP decisions require human authority (ESCALATE_HUMAN).
  • Compliance corrections inside envelope proceed via automatic remediation (AUTO_REMEDIATE_AND_REVALIDATE). Prohibited conditions produce STOP.

Canonical References

  • Shared decision model, gate contract, and ownership matrix: see the governance protocol above.
  • On-demand methods: AUTHORITATIVE_SOURCE_VERIFICATION_GUIDE.md, LICENSE_PRIVACY_IP_COMPLIANCE_GUIDE.md, HUMAN_ESCALATION_BOUNDARIES_GUIDE.md, and examples/governed-change-review-example.md

Token Efficiency

Default to compact output and load only relevant sections. Expand when findings exist.

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.