Works with the AI tools you already use
Claude Code
Cursor
Codex CLI
GitHub Copilot
Gemini CLI+20 moreClaude CodeCursorCodex CLIGitHub CopilotGemini CLI
OpenClaw
Manus
Grok+17 agentscontext budget intelligence evaluator
Audit AI skill token efficiency and map peak context footprint to prevent reasoning-quality degradation.
$9.99
context budget intelligence evaluator
Example session with this skill installed
Audit the context footprint for my new technical-writer skill. It has a 2KB main body and four 10KB reference files. I want to see if it works on 32K models.
- Read your context and instructions
- Compiled the context budget intelligence
- Generated the document
The analysis of your technical-writer skill is complete. While the total package size is ~42KB, the peak footprint is significantly lower due to your file routing structure.
Context Budget Report
Skill: technical-writer
Always-resident (frontmatter description): ~45 tokens
Main body: ~512 tokens
Reference files:
ref/api-style.md: ~2,560 tokensref/sovereign-docs.md: ~2,560 tokensref/formatting-rules.md: ~2,560 tokensref/glossary.md: ~2,560 tokens
Trigger Path Analysis
| Task | Files Loaded | Path Total (tokens) |
|---|---|---|
| API Documentation | body, api-style.md, glossary.md | ~5,677 |
| Style Audit | body, sovereign-docs.md, formatting-rules.md | ~5,677 |
| General Drafting | body, formatting-rules.md | ~3,117 |
Peak Footprint: ~5,677 tokens (from: API Documentation path)
Ceiling Comparison
| Ceiling | Peak Footprint | % Consumed | Flag |
|---|---|---|---|
| 8K (Edge) | 5,677 | 71% | Risk |
| 32K (Standard) | 5,677 | 18% | OK |
| 128K (Large) | 5,677 | 4% | OK |
Structural Bloat
None found. Individual files are well-sized at ~2.5K tokens each.
Claim Consistency
Claims match measured footprint. The skill is safe for 32K models with ample reasoning headroom.
Recommendations
- For 8K models, the "API Documentation" path is tight. Consider splitting
glossary.mdinto domain-specific files to reduce the baseline load for that path. - Add explicit pointers in the main body to ensure the model doesn't pull all four references if the task is ambiguous.
Next steps
- Implement domain-specific glossary splits for smaller context targets.
- Run a side-by-side test on an 8K model to check for hallucinations in API docs.
- Update your README to confirm 32K compatibility.
context-budget-intelligence-evaluator.pdf
PDF · document
Example file from a real run - the skill writes it into your workspace.
Connects securely to your tools. The creator never sees your data.
What you get
About this skill
The problem
Total file size is a poor metric for AI skill performance. The real issue is whether the worst-case simultaneous context load leaves enough headroom for the model to reason effectively without hitting token limits or degrading output quality.
What it does
- Inventories token counts for frontmatter, main body, and reference files individually.
- Maps specific trigger paths to determine exactly which files load for different tasks.
- Calculates peak context footprint based on the single most demanding task path.
- Benchmarks footprint against 8K, 32K, and 128K ceilings to identify reasoning-quality risks.
- Audits "lightweight" claims against actual measured data to ensure marketplace honesty.
Why this beats prompting it yourself
Most LLMs default to summing the entire directory, which overestimates context usage for skills using progressive disclosure. This tool models actual routing logic to find the real peak load and identifies specific structural bloat that simple token counters miss.
Use cases
- Audit a complex skill package before marketplace submission.
- Test if a high-performance prompt can be ported to 8K-context edge models.
- Identify which specific reference files are causing reasoning degradation.
- Verify that a "minimalist" skill isn't secretly bloating the context window.
Known limitations
Provides approximate token counts based on character heuristics. It identifies context-loading risk but cannot confirm actual reasoning degradation without user-provided side-by-side outputs.
How to install
Works the same in every agent - Claude, Cursor, Codex, Copilot and 20+ more.
- 1
Download the ZIP
Free skills download straight away. Paid skills unlock right after purchase.
- 2
Unzip into your skills folder
Every agent reads skills from one folder on your machine. Drop the unzipped folder in there.
- 3
Ask your agent to use it
Restart the agent if it was already running. It picks the skill up automatically - no config needed.
Skills folder by agent
Click the path to copy it. Create the folder if it does not exist yet.
Reviews
No reviews yet
Be one of the first to try it. Every listed skill passes our trust checks below.
Security scanned
Passed our 8-point scan before listing
Fresh listing
Recently published to Agensi
30-day refund
Not a fit? Get your money back
Trust & safety
Security scanned
Verified clean 6 days ago
- Passed all security checks, Safe to install