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 skills/bumblebiber/hmem/hmem-setupnpx skills add Bumblebiber/hmem --skill hmem-setupgit clone --depth 1 https://github.com/Bumblebiber/hmemWhat 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 | $0.00056 | $0.02984 |
| Opus 5 | $0.00028 | $0.01492 |
| Sonnet 5 | $0.00011 | $0.00597 |
| Haiku 4.5 | $0.00006 | $0.00298 |
Grade C, and why
hmem-setup scanned grade C with 2 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 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.
Asks for rootmediumPrivilege escalation
A mod that escalates privileges can change anything on the machine, not only the project.
| Linux (Debian/Ubuntu) | `sudo apt install python3 make g++` | Reads agent configuration directoriesmediumAgent snooping
.claude/, .codex/, .gemini/ hold keys, settings and other credentials a mod has no legitimate need for.
`hmem init` registers 4 hooks in `~/.claude/settings.json`. Each hook is a bash script in `~/.claude/hooks/`. How it starts
The opening of the file, as written. The whole thing — 272 lines — stays where its author put it; the contents beside it link to each section on GitHub.
hmem Setup
Recommended: hmem init
Install hmem globally, then run the interactive installer:
npm install -g hmem
hmem init
hmem init performs all setup steps automatically:
- Detects installed AI tools (Claude Code, Gemini CLI, OpenCode, Cursor, Windsurf, Cline)
- Asks for installation scope (system-wide or project-local)
- Creates the memory directory and optional example database
- Writes
.mcp.jsonwith the correct paths for each detected tool - Adds session-start instructions to the tool's config file (CLAUDE.md, GEMINI.md, etc.)
- Creates
hmem.config.jsonwith sensible defaults - Installs all 4 auto-memory hooks (Claude Code only — see Hook Reference below)
- Copies skill files (slash commands) to the tool's skill directory
After hmem init, install the slash-command skills:
npx hmem update-skills
Restart your AI tool and call read_memory() to verify.
Non-interactive mode (CI / scripting):
hmem init --global --tools claude-code --dir ~/.hmem --no-example
Hook Reference (Claude Code)
hmem init registers 4 hooks in ~/.claude/settings.json. Each hook is a bash script in ~/.claude/hooks/.
1. UserPromptSubmit — memory load + checkpoint reminder
Script: ~/.claude/hooks/hmem-startup.sh
- First message: injects
additionalContexttelling the agent to callread_memory()silently. Also injects H-entries (ID + title; pinned H-entries include their L2 children inline so canonical identity/preference info is visible without extra reads — since v1.3.3), active device apps, infrastructure favorites, recent projects, checkpoint status, and an--- hmem-sync ---block showing link state (✓ linked / ⚠ partial / ✗ not linked / ✗ not configured). Block is always emitted since v1.2.9. - Every Nth message (N =
checkpointInterval, default 20): injects a checkpoint reminder.checkpointMode: "remind"— adds anadditionalContextnudge; the agent decides what to save.checkpointMode: "auto"— checkpoint is handled by the Stop hook instead (no reminder injected).
- Subagents (messages with
parentUuid) are skipped. - Uses a per-session counter file at
/tmp/claude-hmem-counter-{SESSION_ID}.
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 · 272 lines · 56 tokens per session scan C 51136e24c13a
hmem-setup is a skill published in the GitHub repository Bumblebiber/hmem (24 stars, last pushed 1mo ago), licensed MIT. It adds 56 tokens to every session and 2,984 once invoked, about $0.0003 per session on Opus 5. A static security scan graded it C with 2 findings (asks for root, reads agent configuration directories). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.
Other skills, from other repositories
b123d-modeling
Use this skill when asked to model, build, or modify a 3D part or assembly with build123d — from a text description, a technical drawing (image or PDF), dimensions in a spec, or an existing STEP/STL file.
authoring-decorators
Use when the user wants to write a new prompt decorator - either a personal one for their repo or a contribution back to the upstream catalogue. Walks through the JSON schema, the extensions/ convention, how to verify a new decorator end-to-end through the hook, and how to diagnose failures using debug logging and…
b123d-drawing
../../../src/build123dmcp/skills/b123d-drawing/SKILL.md.
mlflow-traces
Use when working with MLflow traces: debugging via MCP tools, analyzing performance, logging feedback, writing custom scorers/evaluations, or cleaning up trace data.
gemini-visualize
Generates interactive HTML visualizations (concept maps, evidence networks, knowledge graphs) from Gemini analysis results. Triggers automatically after /gr:video, /gr:research, /gr:analyze.
tts-production
Produces voiceover audio via ElevenLabs TTS API. Activates for TTS generation, voice tuning, audio ducking, or multilingual narration — not for voice AI agents, transcription, or music.