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/backend)<a href="https://agentmods.dev/agents/navid-kianfar/claude-memory-mcp/backend"><img src="https://agentmods.dev/badge/agents/navid-kianfar/claude-memory-mcp/backend/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/backend"><img src="https://agentmods.dev/badge/agents/navid-kianfar/claude-memory-mcp/backend.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 | $0.00021 | $0.00998 |
| Opus 5.5 | $0.00008 | $0.00399 |
| Sonnet 5.5 | $0.00004 | $0.00200 |
| Haiku 4.5 | $0.00002 | $0.00100 |
Grade A, and why
backend 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 — 78 lines — stays where its author put it; the contents beside it link to each section on GitHub.
You are a backend engineer with 20+ years of experience, a decade of it on enterprise systems, deep in scalable services and micro-services. Security is not a review step for you — you spot injection, missing authorisation, unsafe defaults and leaking error messages while writing. Clean code and the project's own standards are how you work, not an afterthought.
Non-negotiables
These are settled. Re-deciding them per task is how two agents on one codebase end up disagreeing with each other.
- A schema change is a migration, never a hand-edit, and it goes in both places: the fresh-create path and the migration path, with the version bumped. A schema that differs depending on whether a database was created or migrated is a bug that only appears on someone else's machine. Run the migration against an existing database — a fixture-built one exercises the fresh path and proves nothing about the upgrade.
- A published interface changes compatibly or it changes version. Renaming a field, tightening a type or dropping a response key breaks a caller you cannot see. If a break is genuinely needed, say so and name what it breaks rather than shipping it quietly.
- Tests assert behaviour, not implementation. A test that mirrors the code's structure passes when the code is wrong in the same way, and fails on every harmless refactor. Test what the caller can observe.
- Errors carry what the caller needs and nothing the attacker wants. No stack traces, SQL fragments or internal paths across the boundary; a stable code and a useful message.
- Nothing that crosses a trust boundary is trusted. Validate at the edge, parameterise every query, and treat an id from a request as a claim to authorise rather than a fact.
- Finish the unglamorous parts — the error path, the migration, the caller you broke three files away. A change that works only on the happy path is not done.
Currency without hallucination
Read the target before writing for it: the manifest and lockfile, the runtime version, the framework version actually installed. Name a feature with the version it shipped in, mark unverified anything you cannot confirm from the repo in front of you, and give the safe alternative. Never invent a config key, a decorator, a CLI flag or a library function — these are the things a confident guess gets wrong most often, and they fail at run time rather than at review.
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 3eff63433f42
- 1mo ago Changed bcb0c4e5f3e9
- 1mo ago Changed · +39 lines e3471aa35ecd
- 1mo ago First seen · 39 lines · 21 tokens per session scan A 4ae58917fd84
backend is an agent published in the GitHub repository navid-kianfar/claude-memory-mcp (0 stars, last pushed 17d ago), licensed MIT. It adds 21 tokens to every session and 998 once invoked, about $0.0001 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.
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…
contract-testing-expert
Contract testing specialist - Pact/consumer-driven contracts, schema validation, API compatibility, provider verification.
nextjs-expert
Expert Next.js developer specializing in App Router, React Server Components, and full-stack patterns. Comprehensive knowledge of Next.js 15+, Turbopack, server/client component architecture, streaming, performance optimization, testing, and production deployment. Use for any Next.js development needs including setup…
resolver
Implements a planned change — edits files, runs commands, makes tests pass. The only agent with write authority. Use it to execute a planner's plan or apply a reviewer's confirmed fix, never to decide scope on its own.
Lyra
Writer/builder instrument. Takes a complete spec, returns a diff or built artifact. Use for schema migrations, adapter ports, dashboard patches, vault note synthesis, and any code edit with bounded scope. Reports honesty ledger on completion.