Skills DirectorySkills Directory
SkillsLearnSecurityCategoriesDocsCommunityBlog
Sign InSubmit Skill
Skills Directory

Security-tested agent skills for Claude, coding agents, and AI workflows.

Directory

  • Browse Skills
  • All Skills A–Z
  • Claude Skills
  • Claude Code Skills
  • Agent Skills
  • Categories
  • Authors
  • Submit a Skill

Learn

  • Learn Hub
  • Install Claude Skills
  • Write SKILL.md
  • Skills vs MCP
  • Directories Compared

Security

  • Security
  • Methodology
  • Secure Claude Skills
  • Security Badges

Company

  • About
  • Community
  • Blog
  • API Docs
  • Advertise

2026 Skills Directory. All rights reserved.

ProTermsPrivacyRefunds
Back to skills

Idd Gap Check

ASecurity

Adversarially review one or more IDD Specs for implementation ambiguity, contradictions and unowned impact-surface files before execution. Produces finding reports and gate annotations without rewriting Spec content.

6 stars
0 votes
0 copies
0 views
Added 9/22/2026
development

Security Analysis

A100/100

Scanned 9/22/2026

Install to Claude Code

$npx -y skills add GrillerGeek/idd-framework --skill idd-gap-check --agent claude-code

Installs into .claude/skills of the current project.

Are you the author of Idd Gap Check?

Add the live security badge to your README — it updates automatically with every re-scan.

Security grade badge for Idd Gap Check
[![Security: A — Skills Directory](https://www.skillsdirectory.com/api/skills/grillergeek-idd-gap-check/badge)](https://www.skillsdirectory.com/skills/grillergeek-idd-gap-check)

More formats (shields.io, HTML) on the badges page.

Download with Pro
Files
SKILL.md
---
name: idd-gap-check
license: Apache-2.0
description: Adversarially review one or more IDD Specs for implementation ambiguity, contradictions and unowned impact-surface files before execution. Produces finding reports and gate annotations without rewriting Spec content.
---

# Gap-check IDD Specs

Read the bundled [Spec contract](references/spec-reference.md) and
[reviewer procedure](references/gap-reviewer.md). Resolve resources relative to
this installed skill, and project artifacts relative to the consuming project.
Use only available native tools; there is no required plugin environment variable,
model vendor, external agent package or globally installed helper.

This entry point is the **orchestrator**. The reviewer is report-only. If native
subagent delegation is available, delegate this bounded review with explicit
selected IDs, project path, installed reference paths and invalidation evidence.
Honor a host adapter's explicit reviewer policy; otherwise use the host's configured
model. If delegation is unavailable, announce sequential orchestration and reviewer
phases and execute the same procedure locally; the reviewer phase must not edit
Specs or grant itself permission to execute. A tool failure is not a clean review.

## Select before writes

- One ID selects one Spec; multiple IDs select that exact set, deduplicated.
- `all` selects Specs with lifecycle ready or review.
- Empty input lists available ready/review Specs and requests selection.

Resolve exact IDs under docs/specs, including descriptive filenames; require one
file per selected internal ID. Report missing/ambiguous/unreadable targets and
continue only unambiguously selected ones. No eligible target means stop without
writes. Parse one YAML document with duplicate keys rejected before converting
values. Use existing safe YAML tooling; if no safe parser is available, stop with
an actionable limitation. Never repair malformed YAML as part of this review.

## Invalidate, then review

For each safely readable selected Spec, replace or insert exactly one spec.gap_check
mapping with status blocked, blockers 1, warnings 0, report null and today's ISO
date. Preserve every other byte, including lifecycle, comments and all five
blocks. Use a parser to identify the mapping; replace only its text span. A whole
file parse/dump can reformat content and is not an acceptable annotation update.
If the document's formatting prevents a safe isolated edit, stop that target and
report the limitation; do not dispatch it with an old passed annotation.

Keep previous reports until a current completed review replaces them. Pass the
explicit selected IDs and successful invalidation to the reviewer. It checks
completeness 1–10, performs adversarial analysis and the required Coverage sweep,
and returns one current result per selected Spec. Human peer review is a separate
fact, never an inference from the review result.

## Consume current results

- Completeness failure returns item numbers, writes no finding report and never
  reads a previous report as its result. Replace the sentinel with blocked,
  blockers equal to the failed-item count, warnings 0 and report null. Leave any
  old report bytes untouched; continue the complete targets.
- Completed review returns ID, canonical docs/reviews/<SPEC-ID>-gap-check.md and
  unresolved counts. Read this invocation's report and verify the summary counts,
  all required finding fields, and mandatory Coverage. Confirm the quoted Spec
  content is still current. Both counts zero gives passed; blockers zero and
  warnings positive gives warnings; any blocker gives blocked. Update only the
  single annotation and preserve all other Spec bytes.
- Errors, interruption, missing/malformed output, count mismatches or an unverified
  current result retain the blocked operational sentinel with null report. Report
  the failure and rerun; old files and their timestamps are not evidence of a new
  successful review. Do not infer a zero from missing results.

With two or more complete Specs, require the reviewer's combined coverage report
and surface ownership conflicts before coordinated execution. Incomplete targets
are named and excluded. With fewer than two complete Specs use single-Spec
coverage and do not claim portfolio review. Canonical reports have fixed per-Spec
names; portfolio names receive a numeric suffix on collision. No lifecycle change
occurs in this skill. A clean result still requires ready and actual human review
before implementation; a report headed PASS with warnings does not authorize it.

Attribution

GrillerGeekGrillerGeek
View sourceMore from GrillerGeek →
SSkills DirectorySkills Directory

Your tool, in front of Claude Code builders.

3 founder slots · $299/mo · GSC-verified traffic · sponsors can never buy grades.

See placements

Is this your skill, or is something wrong with this listing? Request removal or report an issue. Author removals are honored within 72 hours.

Comments (0)

No comments yet. Be the first to comment!

SSkills DirectorySkills Directory

Your tool, in front of Claude Code builders.

3 founder slots · $299/mo · GSC-verified traffic · sponsors can never buy grades.

See placements

Related Skills

Browser Extension Developer

Use this skill when developing or maintaining browser extension code in the `browser/` directory, including Chrome/Firefox/Edge compatibility, content scripts, background scripts, or i18n updates.

284072 votes

Seo Optimizer

SEO optimization with keyword analysis, readability assessment, technical validation, content quality. Use for search rankings, blog posts, content audits, or encountering keyword density, readability scores, meta tags, schema markup errors.

2192 votes

Google Official Seo Guide

Official Google SEO guide covering search optimization, best practices, Search Console, crawling, indexing, and improving website search visibility based on official Google documentation

1862 votes

Tanstack Start

Build a full-stack TanStack Start app on Cloudflare Workers from scratch — SSR, file-based routing, server functions, D1+Drizzle, better-auth, Tailwind v4+shadcn/ui. Use whenever the user mentions TanStack Start, asks to scaffold a full-stack Cloudflare app with SSR, wants an SSR dashboard, or asks for a React 19 + Cloudflare Workers app with file-based routing and server functions — even if they don't name TanStack Start specifically. No template repo — Claude generates every file fresh per ...

9881 votes

Pentest

PTES-aligned adversarial security audit for backend, frontend, and mobile applications. Produces a CVSS-scored Hacker Report with verified PoCs and phased remediation.

5491 votes
View all in development →