feat(agents): add Challenger / Devil's Advocate agent template + fact-checking skill

[tamirdresher]

Summary

Addresses #598 — Devil's Advocate / Challenger agent pattern.

Adds two deliverables:

1. .squad/templates/agents/challenger.md

Generic challenger agent charter template. Teams drop this in as their "Q", "Advocate", "Auditor", or any role name that fits. The template includes:

• Role definition (constructively contrarian, not obstructionist)
• Iterative retrieval pattern (max 3 investigation cycles)
• Ceremony integration — coordinator auto-spawn pattern for architecture decisions and quantified claims
• Spawn prompt template for coordinators

2. .squad/skills/fact-checking/SKILL.md

Fact-checking skill that backs the challenger agent:

• Per-claim verdict table format (✅ Verified / ⚠️ Unverified / ❌ Contradicted)
• Counter-hypothesis methodology
• Auto-trigger guidance (superlatives, percentages, final deliverables)

Prior Art

Field-tested as the "Q" agent across 200+ issues. Concrete catches:

• Inflated token savings: claimed 75–90%, actual 20–55%
• Fabricated config property references
• Misidentified performance bottlenecks

False positive rate ~15% (challenged things that were correct). Still worthwhile — low cost, high signal.

Checklist

• Agent charter template added
• Fact-checking skill added
• Ceremony integration documented
• Example spawn prompt in template
• No platform-specific dependencies (works with any Squad setup)

Closes #598

[bradygaster]

🔍 Squad Team Review

Reviewed by: Procedures (Prompt Architect) — Pass 3 of consensus ceremony
Action: Hold — proposal-first policy applies
Team Consensus: 5/5 AGREE

Review Summary

This PR adds fact-checking capabilities, enabling skills to verify claims and validate information quality.

Process Note

Per our team's documented proposal-first policy (.squad/decisions.md): "Meaningful changes require a proposal in docs/proposals/ before execution." This is a feature-class change that needs a proposal document before merge.

What's Needed

1. Create a proposal in docs/proposals/ describing the skill's purpose, integration points, and expected behavior
2. Retarget this PR from main to dev (our staging branch)
3. Once proposal is reviewed and approved, this PR can proceed

Thank you for the contribution! The implementation looks solid — we just need the process alignment. 🙏

⏸️ Action: HOLD — awaiting proposal + retarget to dev

[bradygaster]

FIDO Quality Review — PR #603: Challenger / Devil's Advocate agent

Verdict: ✅ MERGE (with notes)

Findings

✅ Charter structure: challenger.md follows the charter template well — has Identity, What I Own, How I Work, Boundaries, Iterative Retrieval, Ceremony Integration, Model, and Voice sections. Good alignment with .squad/templates/charter.md.

✅ Fact-checking skill: .squad/skills/fact-checking/SKILL.md has proper frontmatter (name, description, license, metadata block with confidence). Review methodology and output format are clear and practical.

✅ Additive: Creates new files only, no modifications to existing code.

⚠️ File placement (non-blocking concern):

• .squad/skills/fact-checking/SKILL.md — places skill in the project's own .squad/ directory (team state). This is correct for a team-level skill, not a distributable template skill.
• .squad/templates/agents/challenger.md — creates a new directory. This is fine as a team-level agent template.
• Note: if this is intended as a distributable template (for all Squad users), it should go in packages/squad-cli/templates/ instead.

Missing changeset: No .changeset/ file, but since this adds to .squad/ (team state, not published packages), a changeset may not be required.

Target branch: Targets main — acceptable for team state changes.

Good contribution. The Challenger agent concept fills a real gap in the team's quality review workflow.