ai-stance

name	ai-stance
description	Generate and maintain AI_STANCE.md for any uFawkes* repo. Use when onboarding a repo to the uFawkesAI suite, when reviewing AI policy currency, or when a new AI tool is being adopted. Implements DORA AI Capability 1.
license	MIT
compatibility	Claude Code, GitHub Copilot, OpenCode, Cursor, Codex, Gemini CLI
metadata	{"author":"paruff","suite":"uFawkesAI"}

Skill: AI Stance

Load trigger: "load ai-stance skill" > DORA: AI Capability 1 (Clear and communicated AI stance) Token cost: Low

Purpose

Produce and maintain AI_STANCE.md — the document that gives every contributor (human and agent) unambiguous clarity on how AI tools are used in this repo. Ambiguity creates risk; clarity creates psychological safety for effective experimentation.

DORA AI Capabilities Model v2025.1 identifies four required clarity dimensions: expectation of use, organizational support, permitted tools, and role applicability. This skill ensures all four are present.

Three-Bucket Framework

Every AI_STANCE.md must classify all AI tool usage into exactly three buckets. No "it depends" without specifying what it depends on.

Bucket	Definition	Decision rule
Prohibited	Never permitted in this repo, regardless of context	Safety, compliance, or trust boundary violated
Permitted with guardrails	Allowed when stated conditions are met	Default for most AI-assisted development
Allowed	No special conditions required	Low-risk, fully reversible, human-verifiable

Default Stance for uFawkes* Repos

Pre-populated for the fawkes suite. Override per repo as needed.

Prohibited

• Sending PII, secrets, or proprietary infrastructure configs to public AI models
• Committing AI-generated code without running pre-commit hooks
• Bypassing branch protection rules because "the AI said it was fine"
• Using AI to generate security policy or compliance documentation without human expert review
• AI-generated content in Dojo modules without disclosure to learners

Permitted with Guardrails

Use	Guardrail
AI-generated code merged to main	Human review required; at least one test covering the new code
AI-assisted spec / design documents	Discovery brief must exist first (discover agent ran)
Agent sessions modifying infrastructure configs	j-curve-navigation pre-flight check must pass
AI-generated release notes	Human review before publishing
opencode sessions in any uFawkes* repo	Session must load AGENTS.md and relevant skills first
graphify corpus built from internal docs	Corpus must not include files containing secrets or PII

Allowed

• AI-assisted code completion (Copilot, Claude Code) for any file not in Prohibited list
• AI-generated first drafts of blog posts, dev.to articles, LinkedIn posts
• AI-assisted issue triage and labeling
• AI-generated test stubs (human completes and verifies)
• Asking AI tools to explain existing code or documentation

Four Required Clarity Dimensions

The generated AI_STANCE.md must answer all four:

1. Expectation of use — Is AI use expected, encouraged, optional, or discouraged here?
2. Organizational support — What tooling is provided? What training/docs exist?
3. Permitted tools — Exact tool names and versions (not "any LLM")
4. Role applicability — Does this stance apply to agents, humans, or both?

Permitted Tools Reference (uFawkes suite defaults)

Tool	Version / model	Permitted scope
opencode	Latest stable	All repos — primary agentic development tool
Claude (Anthropic API)	claude-sonnet-4-6	All repos — skill and agent authoring, code review
graphify	[confirm variant]	All repos — context corpus building
ponytail	Latest stable	All repos — YAGNI enforcement
GitHub Copilot	Current	IDE completion only

Note: Confirm graphify variant (safishamsi/graphify vs rhanka/graphify) before publishing AI_STANCE.md. They are different tools with the same name.

AI_STANCE.md Template

# AI Stance — [REPO_NAME]

> Last reviewed: YYYY-MM-DD
> Next review due: YYYY-MM-DD (quarterly)
> Owner: [GITHUB_USERNAME]
> Suite: uFawkesAI

## Expectation of Use

AI-assisted development is [expected / encouraged / optional] in this repo.
[One sentence on why — e.g., "We use AI to clear bottlenecks in the product lifecycle,
not to replace human judgment on architecture and security decisions."]

## Organizational Support

- Permitted tools: listed in the table below
- Skill suite: uFawkesAI `.agents/skills/` — load relevant skills before each session
- Questions or policy concerns: file a GitHub issue with label `ai-policy`
- Policy reviews: quarterly (see ai-policy-lifecycle skill)

## Permitted Tools

| Tool           | Model / version   | Scope                               |
| -------------- | ----------------- | ----------------------------------- |
| opencode       | latest            | Agentic development sessions        |
| Claude         | claude-sonnet-4-6 | Skill authoring, review, generation |
| graphify       | [variant]         | Context corpus                      |
| ponytail       | latest            | YAGNI enforcement                   |
| GitHub Copilot | current           | IDE completion                      |

## Three-Bucket Classification

### Prohibited

[Fill from default stance above, plus any repo-specific additions]

### Permitted with Guardrails

[Fill from default stance above, plus any repo-specific additions]

### Allowed

[Fill from default stance above, plus any repo-specific additions]

## Role Applicability

This stance applies to: human contributors AND AI agents (opencode sessions,
GitHub Actions opencode workflow, any automated agent invocation).
Agents must load `ai-stance` skill and verify this document exists before beginning
substantive work.

Session Actions

When this skill is loaded, the agent should:

1. Check whether AI_STANCE.md exists at repo root
2. If missing: generate it using the template above, prompt human for repo-specific overrides
3. If present: check Last reviewed date — warn if >90 days old
4. Verify all four clarity dimensions are present
5. Verify all listed tools are still current (no EOL tools in Permitted list)
6. Output stance-report.json

Output Format

{
  "skill": "ai-stance",
  "status": "present | missing | stale",
  "ai_stance_path": "AI_STANCE.md",
  "last_reviewed": "YYYY-MM-DD",
  "days_since_review": 42,
  "review_overdue": false,
  "clarity_dimensions_present": {
    "expectation_of_use": true,
    "organizational_support": true,
    "permitted_tools": true,
    "role_applicability": true
  },
  "prohibited_count": 5,
  "guardrail_count": 6,
  "allowed_count": 5,
  "warnings": []
}

Sub-Skills

Sub-skill	Purpose
ai-stance/template	Generates AI_STANCE.md from scratch for a new repo
ai-stance/audit	Reviews existing AI_STANCE.md for completeness and currency
ai-stance/diff	Compares two repos' stances to identify inconsistencies across the suite