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 skills add Knuckles-Team/epistemic-graph --skill epistemic-graph-deploygit clone --depth 1 https://github.com/Knuckles-Team/epistemic-graphWrote 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/skills/knuckles-team/epistemic-graph/epistemic-graph-deploy)<a href="https://agentmods.dev/skills/knuckles-team/epistemic-graph/epistemic-graph-deploy"><img src="https://agentmods.dev/badge/skills/knuckles-team/epistemic-graph/epistemic-graph-deploy/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/skills/knuckles-team/epistemic-graph/epistemic-graph-deploy"><img src="https://agentmods.dev/badge/skills/knuckles-team/epistemic-graph/epistemic-graph-deploy.svg" alt="Reviewed on agentmods" width="80" height="20"></a>- NVIDIA SkillSpector pass
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 | $0.00075 | $0.00949 |
| Opus 5 | $0.00037 | $0.00475 |
| Sonnet 5 | $0.00015 | $0.00190 |
| Haiku 4.5 | $0.00007 | $0.00095 |
Grade A, and why
epistemic-graph-deploy 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 9d 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 — 73 lines — stays where its author put it; the contents beside it link to each section on GitHub.
epistemic-graph deploy / promote
The engine binary is bind-mounted from the host into the Swarm task, so deploy =
replace the host binary → restart the service (the new container re-execs it). The KG
lives in a separate snapshot/redb volume and survives the restart. Python clients
(graph-os, messaging) are editable installs of the canonical checkout, so Python
changes ship by merge-to-main + restart — no rebuild. Full runbook +
rationale: docs/deploy/binary-promotion.md.
The one command
# build (node/full tier), atomically stage the binary, restart the engine through the
# manager, watch the first-boot migration, then live-smoke the new query surface:
scripts/promote_engine.sh --build --migrate --restore-health-period --verify
# also restart the Python consumers (they reconnect + pick up merged agent-utilities):
scripts/promote_engine.sh --restart-consumers
What the script does (and why), step by step:
- Build
--features full(production needs finance/quant/datascience/reasoning; aserver-only build silently drops those + redb). A finance-symbol guard refuses a mis-built binary. - Backup + atomic stage —
cpthe live binary to.bak-<ts>, write the new one to a same-dir temp,mv -fover the target (atomic rename; the running engine keeps its old inode — noETXTBSY; the next start mmaps the new one). - Restart via the manager —
ssh <manager-node> docker service update --update-order stop-first --force.stop-firstbecause the engine binds a single UDS socket (start-first can't bind twice). This node is a worker; service commands run on the manager. - Migration-aware (
--migrate) — extends the healthcheck--health-start-periodso the one-time.mp→redb first-boot migration (which binds the socket only when done) isn't killed into a restart loop; tails the log untilListening on UDS.--restore-health-periodthen resets it to 20s on the now-fast restart. --verify— live UQL smoke (AS OF,RERANK, a parser-reject) proving the new surface serves, not just that the socket bound.
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.
- 9d ago First seen · 73 lines · 75 tokens per session scan A 3dbd8e78a32f
epistemic-graph-deploy is a skill published in the GitHub repository Knuckles-Team/epistemic-graph (10 stars, last pushed 11d ago), licensed MIT. It adds 75 tokens to every session and 949 once invoked, about $0.0004 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 skills, from other repositories
isrvd-ops
Operations instructions for using the isrvd API to deploy containers, manage services and images, configure routes, handle files, and use a web terminal. The instructions include Python, Node.js, and Bash helper scripts.
deployment
Use when taking an app from source to live: choosing the deploy target from requirements (Hetzner+Coolify vs Vercel vs a third), then wiring container → CI → registry → host with build secrets, healthchecks and rollback. NOT one platform's mechanics (that is coolify, vercel, railway, render), NOT the Dockerfile alone…
kubernetes
Kubernetes operations playbook for deploying services: core objects, probes, resource sizing, safe rollouts, and fast kubectl debugging.
lastlight-server
Install and configure a Last Light SERVER — the GitHub maintenance agent plus its docker-compose stack — on a host. Use when the user wants to "set up / install / deploy / stand up a Last Light server or instance", configure its GitHub App, models, managed repos, or domain, or get the agent running for the first time.…
container-manager-multi-context
Operate several container backends and contexts at once — Kubernetes, Docker, Podman, and Swarm — via the container-manager-mcp MCP server's cmmulticontext tool, with per-call backend/context selection and parallel fan-out across a configured pool of contexts. Use when the agent must compare, migrate between, or…
container-manager-swarm
Docker Swarm cluster orchestration via the container-manager-mcp MCP server — initialize/leave a swarm, list nodes and services, and create or remove replicated services across the cluster. Use when the agent must operate a multi-host Swarm from a manager node, inventory cluster services/nodes, or deploy a service. Do…