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 agentmods add agents/dimpagk92/cellar/mastragit clone --depth 1 https://github.com/dimpagk92/cellarWhat 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 | $0.00000 | $0.01598 |
| Opus 5 | $0.00000 | $0.00799 |
| Sonnet 5 | $0.00000 | $0.00320 |
| Haiku 4.5 | $0.00000 | $0.00160 |
Grade A, and why
mastra 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 2d 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.
How it starts
The opening of the file, as written. The whole thing — 149 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Driving CEL from Mastra
This page shows how to drive CEL from a Mastra agent.
Read docs/adapters-cel-agents.md first if you haven't.
Purpose
Mastra is a TypeScript agent framework. Same ecosystem language as Cellar, easy mental model: a Mastra Agent has tools, an instructions prompt, and a model. To drive CEL, you expose cel_see and cel_act as two Mastra tools and let the agent loop.
In this setup:
- Mastra owns the model, tool loop, and stop condition.
- CEL provides perception, action execution, and adapter truth.
- You don't need
cel_think— the Mastra agent is the planner.
Two integration paths:
- Via Mastra's MCP client. Mastra supports MCP tools natively in recent versions; the agent speaks MCP to CEL as a subprocess and imports tools automatically.
- Direct tool-registration. Wrap CEL calls as plain Mastra tools backed by the Node SDK (
@cellar/agent) or raw child-process MCP calls.
Path 2 is shown below because it makes the ownership clearest. Path 1 is more convenient once Mastra's MCP client surface is stable for your version.
TODO: verify with the latest Mastra release notes at https://mastra.ai/docs — the exact MCP-client API has moved between minor versions.
Setup
1. Build CEL and install Mastra in your project
# In your Mastra app
pnpm add @mastra/core @cellar/agent
# or: pnpm add @modelcontextprotocol/sdk # if using raw MCP path
2. Set your LLM credentials
Mastra needs its own model provider credentials (separate from CEL_LLM_*, which only matters if you use cel_think):
export OPENAI_API_KEY=sk-...
# or ANTHROPIC_API_KEY, etc., per Mastra's provider docs
3. Pick the integration path
- MCP client path: register the CEL MCP server in Mastra's MCP config and let Mastra import tools. Consult the current Mastra docs for the exact
createMCPClient/MCPClientconstructor signature. - Direct tool-registration path: call into
@cellar/agentfrom a MastracreateToolfactory. Shown below.
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.
- 2d ago First seen · 149 lines · 0 tokens per session scan A 79ed1df5056f
mastra is an agent published in the GitHub repository dimpagk92/cellar (4 stars, last pushed 23d ago), licensed Apache-2.0. It costs nothing until one of its globs matches a file; then it loads 1,598 tokens. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.
Other agents, from other repositories
accessibility-specialist
Accessibility expert: WCAG 2.2 audits, screen reader compat, keyboard navigation, ARIA patterns, automated a11y testing.
designer
Ornamental design specialist for historical and modern style analysis, colorblind-accessible palettes, and AI-assisted image generation using Z-Image.
hatch3r-ui
UI quality specialist — reviews generated UI for WCAG 2.2 AA conformance, design-token adoption ≥95%, four-state surface contract coverage, and component-library reuse. Use when UI is authored or modified.
hatch3r-ux
UX quality specialist — reviews generated UX flows for error-recovery clarity, first-run success, decisions-per-flow discipline, focus management, and screen-reader announcement. Use when UX flows are authored or modified.
ui-surfaces
The AI summary card and its proposal choreography, the internals panel, the settings tabs, and the sidebar wake queue.
ACCESSIBILITY_REVIEWER
You are an elite Digital Accessibility Auditor holding an IAAP CPWA (Certified Professional in Web Accessibility) certification. Your mandate is to evaluate digital interfaces to ensure genuine, human-centered accessibility, prioritizing functional task success and Universal Design over rote technical compliance.