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/doucs91/hivelore/claude-mdgit clone --depth 1 https://github.com/Doucs91/hiveloreWrote 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/doucs91/hivelore/claude-md)<a href="https://agentmods.dev/instructions/doucs91/hivelore/claude-md"><img src="https://agentmods.dev/badge/instructions/doucs91/hivelore/claude-md.svg" alt="Measured on agentmods" 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 | $0.02349 | $0.02349 |
| Opus 5 | $0.01175 | $0.01175 |
| Sonnet 5 | $0.00470 | $0.00470 |
| Haiku 4.5 | $0.00235 | $0.00235 |
Grade A, and why
hivelore CLAUDE.md scanned grade A with 1 finding 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 3d 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.
Runs shell commandslowCapability
Expected in a hook, worth knowing in a rule or an instructions file.
- **2026-06-07-convention-vscode-cli-calls-through-runhaive** _(applies to: packages/vscode/src/extension.ts, packages/vscode/src/briefingPanel.ts, packages/vscode/src/observabilityProvider.ts)_: Raw child_process in the How it starts
The opening of the file, as written. The whole thing — 105 lines — stays where its author put it; the contents beside it link to each section on GitHub.
See .ai/project-context.md for the full project context.
Memories live under .ai/memories/ (personal/team/module).
Hivelore — mandatory rules for AI agents
Installing or initializing Hivelore in this repository means the team expects AI agents to drive work through Hivelore, not bypass it.
MCP first (default — never optional unless the developer says so)
The Hivelore MCP server (registered under the key haive in this machine's client config — tools appear as mcp__haive__*) must be enabled in your client. Until it is unavailable (misconfiguration report to the developer), you must use it for substantive work:
- Before you edit project source files, configs (
package.json, tooling), or workflows for the current goal, invokeget_briefingwith a short task string (and files / symbols when you know them). If project context was already refreshed in this session and you only need ranked memories, you may callmem_relevant_toinstead—but still consult Hivelore before trusting ungrounded guesses. - Before declaring a coding task finished, invoke the MCP prompt
post_taskwhen the session produced learnings or substantive changes worth capturing.
Do not treat these as nice-to-have: they are default-required for coding agents here.
Session rules
- Immediately when an approach fails (wrong API, wrong pattern, wrong assumption), call
mem_tried— do not wait until the end of the session.
Behavioral triggers — exact situations requiring immediate Hivelore action
These are non-negotiable. Each row is a concrete situation mapped to a specific tool call that must happen in the same response, not at the end of the session.
| Situation | Required action | Tool |
|---|---|---|
| CLI option or API that does not exist | Document before fixing | mem_tried |
| Test fails because of a wrong assumption | Document the assumption | mem_tried |
| Approach took >15 min and had to be redone | Document what failed | mem_tried |
| Choosing A over B for a non-obvious reason | Capture the rationale | mem_save type=decision |
| Discovering surprising library/framework behavior | Capture the trap | mem_save type=gotcha |
| Inventing a pattern used more than once | Capture the convention | mem_save type=convention |
| Completing a task >30 min or >5 files changed | Close with full recap | mem_session_end with discoveries filled |
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.
- 3d ago First seen · 105 lines · 2,349 tokens per session scan A f7d6215c9a41
hivelore CLAUDE.md is an instructions file published in the GitHub repository Doucs91/hivelore (1 stars, last pushed 5d ago), licensed Apache-2.0. It adds 2,349 tokens to every session, about $0.0117 per session on Opus 5. A static security scan graded it A with 1 finding (runs shell commands). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.
Other instructions, from other repositories
memorix CLAUDE.md
Claude Code instructions for AVIDS2/memorix, covering memorix - agent instructions for claude code, using memorix memory tools, when to search memory, when to store memory and when to resolve memory.
memorix GEMINI.md
Gemini CLI instructions for AVIDS2/memorix, covering memorix - cross-agent memory rules, session start - bind project, then load context, during session - capture important context, architecture & decisions and bug fixes & problem solving.
Mem-Forever copilot-instructions.md
Instructions for ilang-ai/Mem-Forever, covering mem-forever, session start, onboarding (soul.md empty), soul.md format and memory update.
Iron-mem CLAUDE.md
Instructions for BMC-INC/Iron-mem, a project described as: Agent-native memory infrastructure for AI coding assistants. Governed session capture, LLM compression, hybrid retrieval (FTS + vector + RRF), temporal knowledge graph, content-addressed storage, compliance-grade classification. Single Rust binary, MCP server.…
fish-bridge-mcp copilot-instructions.md
Instructions for MakeaMouse/fish-bridge-mcp, a project described as: An economical and fuel(token) efficient AI tool, graph session memory for looong chat session.
ai-mind-map copilot-instructions.md
Copilot instructions for shdra06/ai-mind-map, covering ai mind map mcp — code memory engine (v1.4.0), 🚀 first thing to do in every conversation, quick lookup: "i need to..." → use this tool, ⚡ session lifecycle (always use these) and 🔍 find code (instead of grep/reading files).