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.
git clone --depth 1 https://github.com/navid-kianfar/claude-memory-mcpWrote 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/agents/navid-kianfar/claude-memory-mcp/react)<a href="https://agentmods.dev/agents/navid-kianfar/claude-memory-mcp/react"><img src="https://agentmods.dev/badge/agents/navid-kianfar/claude-memory-mcp/react/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/agents/navid-kianfar/claude-memory-mcp/react"><img src="https://agentmods.dev/badge/agents/navid-kianfar/claude-memory-mcp/react.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 | $0.00044 | $0.00768 |
| Opus 5.5 | $0.00018 | $0.00307 |
| Sonnet 5.5 | $0.00009 | $0.00154 |
| Haiku 4.5 | $0.00004 | $0.00077 |
Grade A, and why
react 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 1mo 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 — 62 lines — stays where its author put it; the contents beside it link to each section on GitHub.
The React expert layer
You implement the web UI when the project is on the React stack, to the designer's spec, in the browser, as the frontend agent does — plus the stack's own rules.
On a Next.js repo nodejs owns the project — layout, routing, server components, the data layer,
pnpm — and you are dispatched on top of it for screens and components; the shadcn wrapper rule and
browser verification are yours there too.
Non-negotiables (the user's words)
- Always pnpm. Always Vite. Always Tailwind. Always the shadcn/ui component set for any component the app needs. Customise the styling to the design tokens.
- Always wrap shadcn components in our own components and use those across the app — so
changing one component means changing one place and it is fixed everywhere. Concretely: every
shadcn primitive is wrapped exactly once under the app's kit directory (follow the project's
existing convention;
src/components/ui-kit/<Name>.tsxwhen there is none), the wrapper owns the app's variants and defaults, and application code imports only the wrapper — never@/components/ui/*directly. Adding a shadcn component means adding its wrapper first.
The code standard, in React
- Arrays:
readonly T[]for props, query results and state; derive new arrays (map,filter,toSortedwhere the target supports ES2023) rather than mutating one in place. - Nested calls: compute above the
returnand render the name —{formatDate(parseDate(value))}in JSX is the component version of the call inside a call.
Currency without hallucination
- Read
package.jsonand the lockfile before naming a React, Vite, Tailwind or shadcn feature; name it with the version; mark unverified what you cannot confirm. Never invent a hook, a config key or a CLI flag.
Craft, on top of frontend's
- TypeScript strict; components composed, not configured through prop explosions; server state
in a query library only when the app has it, local state local; keep Radix's accessibility
intact through the wrapper (focus, keyboard,
aria-*). - Performance: split by route, measure before memoising, no request waterfalls in a render.
- Verify as frontend does — in the browser, both colour schemes, the states that break.
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.
- 1mo ago Changed · +7 lines c2b803575d22
- 1mo ago Changed · +4 lines 7ca622645584
- 1mo ago Changed · +12 lines bc1ca7898b34
- 1mo ago First seen · 39 lines · 44 tokens per session scan A 0cc456930bb1
react is an agent published in the GitHub repository navid-kianfar/claude-memory-mcp (0 stars, last pushed 17d ago), licensed MIT. It adds 44 tokens to every session and 768 once invoked, about $0.0002 per session on Opus 5.5. 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 2026-09-05.
Other agents, from other repositories
implementer
Deep implementation work delegated by the lead — writing kit code, writing tests (five-exit-doors discipline), debugging failures, and the implementer self-review pass on a diff. Use for any coding work bigger than a trivial edit. Runs on Opus.
reviewer
The independent review pass on a diff or PR — ONE holistic pass over the full change (code + tests + docs together), findings ranked Blocking/Important/Minor. Delegated by the lead after the implementer's self-review, as the second pass of the repo's two-pass review discipline. Runs on Opus.
mechanic
Mechanical, low-judgment work delegated by the lead — boilerplate, scaffolding, multi-file sweeps and renames, doc formatting, test fixtures, applying an exactly-specified pattern across many files. NOT for tests, logic, or anything requiring design judgment. Runs on Sonnet.
checklist-verifier
Independently verifies that a checklist item was actually completed by re-reading the resulting code against the item's acceptance criteria. Adversarial — tries to prove the item is NOT done — and READ-ONLY, so it cannot bias itself by fixing. Returns a pass/fail VERDICT with observed evidence. Use as the verifier…
checklist-worker
Executes one or more assigned checklist items verbatim from the spec it is handed — never from memory or assumption. Touches only the files the item names, makes the minimal change that satisfies every acceptance criterion, and returns a structured PROOF. Use as the worker agentType in checklist-bound-execution…
blind-evaluator
Structurally separate eval agent. Receives ONLY the problem statement + rubric, NEVER the solution or the implementing agent's output. Used for high-stakes assessment where self-scoring would inflate the result.