القائمة
Audit auto-loaded context budget, detect redundancy, and recommend optimizations.
يُحدَّد Prompt الذي يراجع المصدر أولًا بشكل افتراضي. يمكنك التبديل إلى أمر مباشر أو تنزيل نسخة محلية.
اقرأ SKILL.md وأي ملفات مرافقة يعرضها SkillsMP قبل أن تقرر التثبيت.
التثبيت باستخدام Codex أو Claude انسخ هذا Prompt والصقه في Codex أو Claude أو مساعد آخر ليراجع صفحة Skill ويثبّتها لك.
يتجاوز الأمر المباشر Prompt المخصّص للمراجعة. افحص المصدر قبل تشغيله.
npx skills add https://github.com/coreindustries/core-ai-template --skill contextيبقى الأمر في سطر واحد. مرّر أفقيًا لمراجعته كاملًا قبل النسخ.
تفضّل نسخة محلية؟ نزّل الملفات المتاحة حاليًا لدى SkillsMP.
Write a structured handoff at session end. Preserves context so the next agent can resume without human briefing. Invoke before ending any feature session longer than 30 minutes.
Multi-perspective code review against project standards with P1/P2/P3 severity classification. Works in Claude Code (Agent + optional GitHub MCP) and Cursor (Task subagents + gh/git). Use when the user invokes /review, asks for a PR or diff review, or wants a standards-aligned review with severity tags.
Multi-perspective code review (P1/P2/P3) for Cursor: inline checklists plus three parallel Task subagents (perf-auditor, security-reviewer, simplicity-reviewer with combined data-integrity prompt). Use when the user invokes /review, asks for a PR review, or wants repo-standard findings with severity.
عرض SKILL.md
استنادا إلى تصنيف SOC المهني
| name | context |
| description | Audit auto-loaded context budget, detect redundancy, and recommend optimizations. |
Audit auto-loaded context budget, detect redundancy, and recommend optimizations.
/context [action] [--verbose]
action: audit, redundancy, recommendations, compare (default: audit)--verbose: Show per-line token estimates for each fileWhen this skill is invoked:
Autonomy:
Token Estimation:
/context audit)Measure all auto-loaded files and report token usage:
CLAUDE.md — count characters, estimate tokens.claude/rules/*.md — read each file, estimate tokensrules/ that point to rules-available/Report format:
Context Budget Report
=====================
Auto-loaded files:
CLAUDE.md 1,265 tokens
rules/ai-agent-patterns.md 1,212 tokens
rules/code-quality.md 842 tokens
rules/error-handling.md 534 tokens
rules/git-workflow.md 687 tokens
rules/quality-checks.md 612 tokens
rules/security-core.md 1,705 tokens
rules/task-management.md 786 tokens
─────────────────────────────────────────────
Total: 7,643 tokens (3.8% of 200K)
Platform rules (enabled):
(none active)
Available platform rules (not enabled):
rules-available/android.md ~XXX tokens
rules-available/docker.md ~XXX tokens
rules-available/ios.md ~XXX tokens
rules-available/nextjs.md ~XXX tokens
rules-available/python.md ~XXX tokens
rules-available/security-mobile.md ~XXX tokens
rules-available/security-owasp.md ~XXX tokens
rules-available/security-web.md ~XXX tokens
/context redundancy)Cross-reference auto-loaded files for duplicated content:
{placeholder} references from each auto-loaded fileReport format:
Redundancy Findings
===================
Shared placeholders (appear in 2+ auto-loaded files):
{lint_fix} → CLAUDE.md:48, quality-checks.md:45
{type_check} → CLAUDE.md:49, quality-checks.md:39
{test_with_coverage} → CLAUDE.md:50, quality-checks.md:47
Duplicate guidance:
- CLAUDE.md "Context Recovery" section overlaps with /resume skill
(low priority — CLAUDE.md version is a brief pointer)
Summary: 3 shared placeholders, 0 high-priority duplicates
/context recommendations)Produce prioritized optimization suggestions:
Report format:
Context Optimization Recommendations
=====================================
P1 - High Impact:
- security-core.md (1,705 tokens): Consider moving detailed tables
and checklists to .claude/references/security-checklist.md.
Auto-loaded portion could be a ~400 token summary with
"see references/security-checklist.md for details".
Estimated savings: ~1,200 tokens
P2 - Medium Impact:
- {lint_fix} placeholder appears in 2 auto-loaded files.
CLAUDE.md Commands section duplicates quality-checks.md.
Consider removing command block from one location.
Estimated savings: ~150 tokens
P3 - Low Impact:
- quality-checks.md checklist section could move to references.
Estimated savings: ~80 tokens
Total potential savings: ~1,430 tokens (18.7% of current budget)
/context compare)Show the budget impact of enabling/disabling platform rules:
rules-available/ — estimate tokens for eachrules/)Report format:
Platform Rule Budget Comparison
===============================
Current baseline: 7,643 tokens (3.8% of 200K)
If enabled:
+ python.md +892 tokens → 8,535 (4.3%)
+ security-owasp.md +1,105 tokens → 8,748 (4.4%)
+ nextjs.md +1,340 tokens → 8,983 (4.5%)
+ docker.md +654 tokens → 8,297 (4.1%)
+ ios.md +780 tokens → 8,423 (4.2%)
+ android.md +720 tokens → 8,363 (4.2%)
+ security-web.md +940 tokens → 8,583 (4.3%)
+ security-mobile.md +830 tokens → 8,473 (4.2%)
Common combinations:
make enable-web (nextjs + security-web + security-owasp)
+3,385 tokens → 11,028 (5.5%)
make enable-python (python + security-owasp)
+1,997 tokens → 9,640 (4.8%)
make enable-mobile (security-mobile + security-web + security-owasp)
+2,875 tokens → 10,518 (5.3%)
content.length charactersMath.round(charCount / 4) — acceptable approximationls -la .claude/rules/ via Bash to identify symlinks pointing to rules-available/These mappings come from the Makefile enable-* targets:
| Target | Rules Enabled |
|---|---|
make enable-web | nextjs, security-web, security-owasp |
make enable-python | python, security-owasp |
make enable-api | security-owasp |
make enable-ios | ios, security-owasp |
make enable-android | android, security-owasp |
make enable-mobile | security-mobile, security-web, security-owasp |
make enable-docker | docker, security-owasp |
$ /context
Context Budget Report
=====================
Auto-loaded files:
CLAUDE.md 1,265 tokens
rules/ai-agent-patterns.md 1,212 tokens
rules/code-quality.md 842 tokens
rules/error-handling.md 534 tokens
rules/git-workflow.md 687 tokens
rules/quality-checks.md 612 tokens
rules/security-core.md 1,705 tokens
rules/task-management.md 786 tokens
─────────────────────────────────────────────
Total: 7,643 tokens (3.8% of 200K)
Platform rules (enabled):
(none active)
Redundancy findings:
3 shared placeholders across auto-loaded files
0 high-priority content duplicates
Top recommendation:
P1: Move security-core.md details to references (~1,200 token savings)
Run '/context recommendations' for full optimization plan.
Run '/context compare' to see platform rule budget impact.