Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx skills add zgsm-ai/costrict --skill evals-contextgit clone --depth 1 https://github.com/zgsm-ai/costrictWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/skills/zgsm-ai/costrict/evals-context)<a href="https://agentmods.dev/skills/zgsm-ai/costrict/evals-context"><img src="https://agentmods.dev/badge/skills/zgsm-ai/costrict/evals-context/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/skills/zgsm-ai/costrict/evals-context"><img src="https://agentmods.dev/badge/skills/zgsm-ai/costrict/evals-context.svg" alt="Reviewed on agentmods" width="80" height="20"></a>What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00084 | $0.01856 |
| Opus 5 | $0.00042 | $0.00928 |
| Sonnet 5 | $0.00017 | $0.00371 |
| Haiku 4.5 | $0.00008 | $0.00186 |
Grade A, and why
evals-context scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 9d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
This is a copy
98% identical to evals-context — 4 lines differ, which has more behind it and is treated as the original. This page carries a canonical link to it rather than competing with it.
How it starts
The opening of the file, as written. The whole thing — 189 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Evals Codebase Context
When to Use This Skill
Use this skill when the task involves:
- Modifying or debugging the evals execution infrastructure
- Adding new eval exercises or languages
- Working with the evals web interface (apps/web-evals)
- Modifying the public evals display page on roocode.com
- Understanding where evals code lives in this monorepo
When NOT to Use This Skill
Do NOT use this skill when:
- Working on unrelated parts of the codebase (extension, webview-ui, etc.)
- The task is purely about the VS Code extension's core functionality
- Working on the main website pages that don't involve evals
Key Disambiguation: Two "Evals" Locations
This monorepo has two distinct evals-related locations that can cause confusion:
| Component | Path | Purpose |
|---|---|---|
| Evals Execution System | packages/evals/ |
Core eval infrastructure: CLI, DB schema, Docker configs |
| Evals Management UI | apps/web-evals/ |
Next.js app for creating/monitoring eval runs (localhost:3446) |
| Website Evals Page | apps/web-roo-code/src/app/evals/ |
Public roocode.com page displaying eval results |
| External Exercises Repo | Roo-Code-Evals | Actual coding exercises (NOT in this monorepo) |
Directory Structure Reference
packages/evals/ - Core Evals Package
packages/evals/
├── ARCHITECTURE.md # Detailed architecture documentation
├── ADDING-EVALS.md # Guide for adding new exercises/languages
├── README.md # Setup and running instructions
├── docker-compose.yml # Container orchestration
├── Dockerfile.runner # Runner container definition
├── Dockerfile.web # Web app container
├── drizzle.config.ts # Database ORM config
├── src/
│ ├── index.ts # Package exports
│ ├── cli/ # CLI commands for running evals
│ │ ├── runEvals.ts # Orchestrates complete eval runs
│ │ ├── runTask.ts # Executes individual tasks in containers
│ │ ├── runUnitTest.ts # Validates task completion via tests
│ │ └── redis.ts # Redis pub/sub integration
│ ├── db/
│ │ ├── schema.ts # Database schema (runs, tasks)
│ │ ├── queries/ # Database query functions
│ │ └── migrations/ # SQL migrations
│ └── exercises/
│ └── index.ts # Exercise loading utilities
└── scripts/
└── setup.sh # Local macOS setup script
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 9d ago First seen · 189 lines · 84 tokens per session scan A edd86e15cc74
evals-context is a skill published in the GitHub repository zgsm-ai/costrict (4,406 stars, last pushed 21d ago), licensed Apache-2.0. It adds 84 tokens to every session and 1,856 once invoked, about $0.0004 per session on Opus 5. A static security scan graded it A with 0 findings. It is 98% identical to evals-context, differing in 4 lines, and is treated as a copy.
Other skills, from other repositories
design-taste-frontend
Anti-slop frontend skill for landing pages, portfolios, and redesigns. The agent reads the brief, infers the right design direction, and ships interfaces that do not look templated. Real design systems when applicable, audit-first on redesigns, strict pre-flight check.
image-to-code
Elite website image-to-code skill for Codex. For visually important web tasks, it must first generate the design image(s) itself, deeply analyze them, then implement the website to match them as closely as possible. In Codex, it must prefer large, readable, section-specific images instead of tiny compressed boards…
brandkit
Premium brand-kit image generation skill for creating high-end brand-guidelines boards, logo systems, identity decks, and visual-world presentations. Trained for minimalist, cinematic, editorial, dark-tech, luxury, cultural, security, gaming, developer-tool, and consumer-app brand systems. Optimized for intentional…
redesign-existing-projects
Upgrades existing websites and apps to premium quality. Audits current design, identifies generic AI patterns, and applies high-end design standards without breaking functionality. Works with any CSS framework or vanilla CSS.
gpt-taste
Elite UX/UI & Advanced GSAP Motion Engineer. Enforces Python-driven true randomization for layout variance, strict AIDA page structure, wide editorial typography (bans 6-line wraps), gapless bento grids, strict GSAP ScrollTriggers (pinning, stacking, scrubbing), inline micro-images, and massive section spacing.
minimalist-ui
Clean editorial-style interfaces. Warm monochrome palette, typographic contrast, flat bento grids, muted pastels. No gradients, no heavy shadows.