Borrowing it
Nothing to install: this file belongs to kitfunso/hippo-memory. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.
curl -O https://raw.githubusercontent.com/kitfunso/hippo-memory/master/AGENTS.mdgit clone --depth 1 https://github.com/kitfunso/hippo-memoryWrote 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/instructions/kitfunso/hippo-memory/agents-md)<a href="https://agentmods.dev/instructions/kitfunso/hippo-memory/agents-md"><img src="https://agentmods.dev/badge/instructions/kitfunso/hippo-memory/agents-md/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/instructions/kitfunso/hippo-memory/agents-md"><img src="https://agentmods.dev/badge/instructions/kitfunso/hippo-memory/agents-md.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 | $0.01147 | $0.01147 |
| Opus 5.5 | $0.00459 | $0.00459 |
| Sonnet 5.5 | $0.00229 | $0.00229 |
| Haiku 4.5 | $0.00115 | $0.00115 |
Grade A, and why
hippo-memory 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 6d 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 — 71 lines — stays where its author put it; the contents beside it link to each section on GitHub.
AGENTS.md - Hippo
Project
hippo-memory, a TypeScript memory system and CLI with UI, integrations, benchmarks, and evaluation harnesses.- Read
README.md,PLAN.md,ROADMAP.md, and relevant docs before broad product or architecture changes. - Work plane boundary:
docs/plans/2026-09-12-work-plane-boundary.md(ROADMAP Part VII, Track W, W0).
Commands
npm run build
npm test
npm run test:eval # the eight slow token-eval harness tests, which npm test skips; --project codex runs the four Codex ones
npm run build:ui
npm run build:all
npm run smoke:pack
npm run smoke:openclaw-install
Rules
- Preserve public CLI and package APIs unless the user asks for a breaking change.
- Document public APIs with JSDoc.
- Prefer focused tests in
tests/or nearby integration/eval harnesses before changing memory behavior. - Name a new test file after the behaviour it pins, not the ticket (
store-stats-concurrency.test.ts, nota7-t3.test.ts). Seed test stores throughinitStore,createMemoryandwriteEntry; write raw SQL only when the test is about SQL. - A new test file needs a contract with no owner file: find the owner and add a row first.
- One owner per contract at the strongest boundary, with one wire-up row per other transport; a test written to carry a refactor is deleted by the PR that lands it.
- No production export for a test; seed helpers live in
tests/_helpers. - Do not commit generated
dist/or UI build output unless the repo expects it for a release. - A PR's changelog entry goes in its own
changelog.d/<branch-with-dashes>.md, never inCHANGELOG.md; the release commit folds them in (changelog.d/README.md). - CI fails when any oxlint rule's hit count rises above
.oxlint-baseline.json. Fix new hits in the files you touch; after clearing old ones,node scripts/check-lint-ratchet.mjs --updateand commit the lower baseline. - Comments say why, in one line. Ticket codes, versions, reviewer notes and dates go to docs or the git log; CI fails when a file's count of them rises above
.comment-history-baseline.json. After rewriting old ones,node scripts/check-comment-history.mjs --updateand commit the lower baseline. - No
src/file over 800 lines and no function over 80. CI fails when a new one appears or one in.size-baseline.jsongrows; after shrinking old ones,node scripts/check-size-ratchet.mjs --updateand commit the lower baseline. - An eval that reads host transcripts (
~/.claude/projects/*.jsonl) copies its corpus outside the repo at registration and names the copy in the prereg: Claude Code deletes sessions after 30 days by default, so a live-path corpus cannot be re-run. - Use Hippo memory commands when useful, but do not store secrets.
- The
compactionstable is one record per Claude Code compaction, not a memory row: recall, FTS, context and sleep's memory passes never read it. Memories saved from a compaction carry thecompaction-memorytag and acompaction:source and fade like any other memory; a restated item is skipped, and another session's restatement strengthens the row it repeats. The keep rule (KEEP_PAIRSinsrc/core/memory.ts) covers imported agent memories only and needs tag and source both;canAutoDeleteandAUTO_DELETABLE_SQL(src/store/rule-sql.ts) must change together. No automatic pass (sleep, dedupe, audit) deletes a memory that backs an object:AUTOMATIC_DELETE_SQLinsrc/store/delete-and-batch.tschecks every table inMEMORY_BACKED_TABLES.
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.
- 6d ago Changed · +11 tokens per session ec62d3797e91
- 7d ago Changed · +12 tokens per session afea3c16a28b
- 8d ago Changed · +3 lines · +75 tokens per session 55c977b2c8c0
- 10d ago Changed · +1 lines · +19 tokens per session 8b07917ad521
- 12d ago Changed · +1 lines · +64 tokens per session 28142039eb6b
- 13d ago Changed · +77 tokens per session 13ca21618f5f
- 17d ago Changed · +5 lines · +223 tokens per session fb2083dc61f7
- 20d ago Changed · +1 lines · +55 tokens per session 5e1115e22997
- 23d ago Changed · +2 lines · +115 tokens per session 2415a1e5679e
- 24d ago First seen · 58 lines · 496 tokens per session scan A a2f3a73ebbcf
hippo-memory AGENTS.md is an instructions file published in the GitHub repository kitfunso/hippo-memory (780 stars, last pushed today), licensed MIT. It adds 1,147 tokens to every session, about $0.0046 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 2026-08-30.
Other instructions, from other repositories
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).
next.js AGENTS.md
AGENTS.md instructions for vercel/next.js, covering next.js development guide, codebase structure, monorepo overview, core package: packages/next and other important packages.
vscode oss-third-party-notices.instructions.md
Instructions for microsoft/vscode, covering vs code oss third-party-notices pipeline, architecture, pipeline flow in ci, cache eligibility and rollout and applying the notice (cutover).
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.
deepseek-harness AGENTS.md
AGENTS.md instructions for deepseek-ai/deepseek-harness, covering agents.md, pre-stable apis and released session data, repository layout, commands and host sandbox failures.
vscode copilot-instructions.md
Copilot instructions for microsoft/vscode, covering vs code copilot instructions, project overview, root folders, core architecture (src/ folder) and built-in extensions (extensions/ folder).