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 instructions/docgotham/memsum/agents-mdgit clone --depth 1 https://github.com/docgotham/memsumWhat 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 | $0.03186 | $0.03186 |
| Opus 5 | $0.01593 | $0.01593 |
| Sonnet 5 | $0.00637 | $0.00637 |
| Haiku 4.5 | $0.00319 | $0.00319 |
Grade A, and why
memsum AGENTS.md 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 yesterday.
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 — 175 lines — stays where its author put it; the contents beside it link to each section on GitHub.
AGENTS.md - Mem·Sum
This is the repository-level agent entry point for Mem·Sum.
For vault operations, the operational contract is DMSUM.md. Read it before changing or querying a Mem·Sum vault. In the local test, read the runtime vault copy through MCP:
read_file("DMSUM.md")
Core Orientation
Mem·Sum is an agent-native, relationship-scoped communication and memory layer. It is not the agent. A participant writes to their own agent, usually with +sum, +dm, or +dmsum, and the agent routes the act through the passive Mem·Sum layer.
Identity and routing can be global. Meaning is relational. In routed local mode, one MCP server may use .dmsum/registry.json to route across many relationship workspaces, but each workspace still has its own raw interactions, wiki updates, wiki graph, participants, and participant-specific preferences.
The server protects structure, provenance, IDs, timestamps, paths, coordination, and dry-run notifications. Agents provide the judgment: interpreting ordinary language, choosing target wiki pages, creating supporting wiki pages, enriching resources, linking the graph, and writing useful prose.
The current product direction: a hosted Supabase/Postgres graph exposed through a remote MCP/API on Vercel. Local filesystem and Git-sync work remain useful prototype, export, backup, and development infrastructure, but they are no longer the core product sync architecture.
Current State And Invariants (2026-07-13)
The product is live. The hosted MCP endpoint is https://sum.memsum.ai/mcp (the bare host
https://sum.memsum.ai also serves MCP); the dashboard, site, and invite links are on
https://memsum.ai. The legacy host dmsum-hosted-mvp.vercel.app keeps serving everything, so
connectors registered against it continue to work. The kernel is open source (Apache-2.0) at
https://github.com/docgotham/memsum, published as snapshots from a private operating archive.
Naming invariants — prose says Mem·Sum; these machine-facing names are frozen on purpose and must not be renamed in a cleanup:
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.
- yesterday First seen · 175 lines · 3,186 tokens per session scan A 58ccdda9abe2
memsum AGENTS.md is an instructions file published in the GitHub repository docgotham/memsum (0 stars, last pushed 10d ago), licensed Apache-2.0. It adds 3,186 tokens to every session, about $0.0159 per session on Opus 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 2026-08-31.
Other instructions, from other repositories
spec-kit AGENTS.md
AGENTS.md instructions for github/spec-kit, covering agents.md, about spec kit and specify, quickstart — add a new integration in 5 steps, integration architecture and integrationmanifest — file tracking.
vscode buildNext.instructions.md
Working notes and architecture documentation for the new esbuild-based build system in build/next. Use when making changes to the new build pipeline (transpile/bundle commands, NLS plugin, source-map handling, resource copying, or self-hosting watch tasks).
codex AGENTS.md
AGENTS.md instructions for openai/codex, covering rust/codex-rs, the codex-core crate, code review rules, crate api surface and model visible context.
langchain AGENTS.md
AGENTS.md instructions for langchain-ai/langchain, covering global development guidelines for the langchain monorepo, corridor security analysis, project architecture and context, monorepo structure and development tools & commands.
vscode oss-third-party-notices.instructions.md
Instructions for microsoft/vscode, covering vs code oss third-party-notices pipeline, architecture, pipeline flow in ci, applying the notice (cutover) and fallback chain (never fail the build).
next.js AGENTS.md
Instructions for vercel/next.js, covering next.js development guide, codebase structure, monorepo overview, core package: packages/next and other important packages.