Skill

folio:review

IP/redline scan, hostile review (white paper + hybrid), standard review with scorecard, and repair loop (max 3 rounds).

npx claudepluginhub athan-dial/skills --plugin folio

Tool Access

This skill uses the workspace's default tool permissions.

Preview

Run IP scanning, mode-appropriate review passes, scorecard generation, and repair loop.

SKILL.md

Similar Skills

cache-components

139.2k

Guides Next.js Cache Components and Partial Prerendering (PPR) with cacheComponents enabled. Implements 'use cache', cacheLife(), cacheTag(), revalidateTag(), static/dynamic optimization, and cache debugging.

cache-components

mcp-builder

124.2k

Guides building MCP servers enabling LLMs to interact with external services via tools. Covers best practices, TypeScript/Node (MCP SDK), Python (FastMCP).

9 files

anthropics-skills-13

canvas-design

124.2k

Generates original PNG/PDF visual art via design philosophy manifestos for posters, graphics, and static designs on user request.

20 files

anthropics-skills-13

Stats

Parent Repo Stars0

Parent Repo Forks0

Last CommitApr 15, 2026

Actions

View Source View Plugin View on GitHub View README

folio:review — Review and Repair

Run IP scanning, mode-appropriate review passes, scorecard generation, and repair loop.

Entry Conditions

Draft exists: drafts/paper.md (white paper/hybrid) and/or drafts/paper.tex (research paper/hybrid).
Read route.json for effective mode.

If draft is missing, direct user to run folio:draft first.

Protocol

1. IP Scan (All Modes)

python ../../scripts/scan_redlines.py workspace/

Writes reviews/ip_safety_report.md.

IP Gate: If violations found, BLOCK — do not proceed until cleared, redacted, or explicitly acknowledged by user. Log acknowledgments in logs/run_log.md. See ../../references/gates.md for IP gate failure handling.

2. Hostile Review (White Paper + Hybrid Only)

Read drafts/paper.md as an external critic. Write reviews/hostile_review.md:

Logical gaps
Unsupported assertions
Competitive risk
Clarity issues
Factual risk

For research paper mode, skip this step.

3. Standard Review (All Modes)

Write reviews/review_round_1.md:

Argument flow, claim support, citations, figures/tables, clarity, venue fit, line-level suggestions.

Write reviews/scorecard.json with dimensions:

argument_coherence
evidence_support
citation_quality
writing_clarity
venue_fit
overall (numeric, consistent scale across rounds)
blocking_issues (list of strings; empty when clear)
suggestions

4. Repair Loop

If blocking issues exist:

Apply fixes to draft.
Save revised draft.
Re-score with new reviews/scorecard.json.

Gate G — Quality non-regression: If overall drops vs. prior best, revert to the better draft.

Max 3 review rounds. After 3 rounds, keep the best-scoring artifact and document residual issues.

5. Hybrid Mode: Dual Review

For hybrid, both tracks must pass:

Conceptual parts: hostile review gate.
Evidence parts: standard review + scorecard gate. No "average failure" — fix or defer with user sign-off.

Exit Conditions

At least one review round completed.
Best draft selected.
Blocking issues resolved or explicitly deferred by user.
IP scan clean or violations acknowledged.
Gate G passes.

Next Stage

Proceed to folio:export.