Add source-grounded evidence map skill 🤖🤖🤖

[alsoleg89]

Summary

• add a build-evidence-map Agent Skill for contested technical choices, research synthesis, and consequential decisions
• preserve supporting, contradicting, qualifying, and missing evidence as typed edges connected to exact source regions
• bundle the zero-dependency evidence contract and offline validator so Copilot can produce a real receipt without npm or network access
• add the schema and evidence-quality references and regenerate the skills catalog entry

Why

Research output often collapses disagreement into polished prose. This skill keeps the provisional position, counterevidence, scope qualifications, and decision-changing unknowns inspectable in a portable .doubt.json artifact.

The bundled validator is intentionally fail-closed. A file hash, node count, JSON parse, or manual schema review cannot be reported as successful validation. This closes a failure observed in a controlled Codex portability run where a network-blocked package command failed but the agent still claimed success. The before/after artifacts and sanitized sessions are public in alsoleg89/doubt.

Validation

• npm run skill:validate — all 396 skill folders valid
• npm run build — catalog entry includes both references and bundled scripts
• bundled validator run against the canonical Agent Skills vs MCP map — VALID 621b1c85104757e4b3752764710d07fb9d6f721c26eaa3e74ee9b1563d82ad89
• bundled contract is byte-identical to the released doubt-ai@0.8.0 contract
• git diff --check

Source retrieval remains explicit: the skill never performs network verification without permission, and it does not describe structural validity as proof that a source or conclusion is true.

[github-actions]

🟡 Contributor Reputation Check: MEDIUM risk

Check	Risk
Profile	MEDIUM
Credential audit	NONE

Maintainers: please review this contributor before merging.
See the workflow run for full details.
Automated check powered by AGT.

[github-actions]

🔒 PR Risk Scan Results

Scanned 5 changed file(s).

Severity	Count
🔴 High	0
🟠 Medium	0
ℹ️ Info	2

Severity	Rule	File	Line	Match
ℹ️	skill-script-touched	skills/build-evidence-map/scripts/contract.mjs	1	skills/build-evidence-map/scripts/contract.mjs
ℹ️	skill-script-touched	skills/build-evidence-map/scripts/validate.mjs	1	skills/build-evidence-map/scripts/validate.mjs

This is an automated soft-gate report. Findings indicate review targets and do not block merge by themselves.

[github-actions]

🔍 Vally Lint Results

✅ All checks passed

Scope	Checked
Skills	1
Agents	0
Total	1

Severity	Count
❌ Errors	0
⚠️ Warnings	0
ℹ️ Advisories	0

Summary

Level	Finding
ℹ️	✅ build-evidence-map (2/2 checks passed)
ℹ️	✓ [spec-compliance] All 1 skill(s) are spec-compliant.
ℹ️	✓ spec-compliance: All spec checks passed.
ℹ️	✓ [valid-refs] All file references across 1 skill(s) are valid.
ℹ️	✓ valid-refs: All file references resolve to existing files within the skill directory.
ℹ️	1 skill(s) linted, 1 passed

Full linter output

### Linting skills/build-evidence-map
✅ build-evidence-map (2/2 checks passed)
    ✓ [spec-compliance] All 1 skill(s) are spec-compliant.
        ✓ spec-compliance: All spec checks passed.
    ✓ [valid-refs] All file references across 1 skill(s) are valid.
        ✓ valid-refs: All file references resolve to existing files within the skill directory.

1 skill(s) linted, 1 passed