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-troubleshootinggit 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-troubleshooting)<a href="https://agentmods.dev/skills/knuckles-team/epistemic-graph/epistemic-graph-troubleshooting"><img src="https://agentmods.dev/badge/skills/knuckles-team/epistemic-graph/epistemic-graph-troubleshooting/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-troubleshooting"><img src="https://agentmods.dev/badge/skills/knuckles-team/epistemic-graph/epistemic-graph-troubleshooting.svg" alt="Reviewed on agentmods" width="80" height="20"></a>- NVIDIA SkillSpector warn
SkillSpector: 1 finding, up to high
These are SkillSpector’s own severities. On a checked sample its high-severity flags on skills were ~96% false positives — a documented command, a public API, a “never do X” rule — so we show them as a caution to read, not a verdict. Why →
- high Tool Misuse · line 78 Tool calls are chained to bypass individual safety checks or escalate capabilities beyond what any single tool call would allow.Fix: Limit tool chaining depth and validate the output of each tool before passing it to the next. Require explicit user approval for multi-step chains.
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 | $0.00082 | $0.01396 |
| Opus 5 | $0.00041 | $0.00698 |
| Sonnet 5 | $0.00016 | $0.00279 |
| Haiku 4.5 | $0.00008 | $0.00140 |
Grade A, and why
epistemic-graph-troubleshooting scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 11d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 103 lines — stays where its author put it; the contents beside it link to each section on GitHub.
epistemic-graph troubleshooting
A runbook for an AI agent operating the engine natively. The engine is a Rust service
on a Unix socket, run as a Docker Swarm service; Python graph-os (MCP) + graph-os-host
(daemon) + messaging are clients of it. Diagnose top-down: engine → socket → clients.
First, triage
docker ps -a --format '{{.Names}}\t{{.Status}}' | grep -E 'epistemic-graph|graph-os|messaging'
ls -la /run/epistemic-graph/epistemic-graph.sock # socket present?
A healthy system = one epistemic-graph_epistemic-graph.* (healthy), one
graph-os_graph-os.* (healthy), one graph-os_graph-os-host.* (healthy). More than one of
any (in the same .N slot) is a duplicate — see below.
Symptom → cause → fix
Engine container down / restart-looping
cid=$(docker ps -aq -f name=epistemic-graph_epistemic-graph | head -1)
docker logs "$cid" 2>&1 | tail -30
Listening on UDSnot yet printed but lock acquired → slow first boot: the snapshot /.mp→redb migration binds the socket only when done (can take minutes on a large KG). Wait; do not kill it. (Deploys: usepromote_engine.sh --migrate.)docker restartfails with "PID … is zombie and can not be killed" → the old process didn't reap children. Usedocker rm -f <cid>; Swarm reschedules a fresh task.- Exit 139 (SIGSEGV) / 137 (OOM-kill) repeatedly → a bad binary or memory pressure.
Roll back the binary (see
epistemic-graph-migrations) and check host memory.
ConnectionRefusedError / [Errno 111] from a client
The socket file may exist but nothing is listening (engine mid-startup) or the engine
restarted and the client hasn't reconnected. Confirm the engine printed Listening on UDS,
then let the client's circuit breaker re-probe (below). The socket lives at
/run/epistemic-graph/epistemic-graph.sock (shared bind mount).
Host daemon crash-loops with "Unknown graph backend type: '…'" / "A persistent graph backend is required"
The deployment's GRAPH_BACKEND env names a removed/unknown backend. tiered is
auto-migrated forward to epistemic_graph (a warning, not a crash) as of the
engine-authority consolidation; any other unknown value still fails. Set the consumer
service env to epistemic_graph (self-contained authority) or fanout (+ mirrors) on the
manager (docker service update --env-add GRAPH_BACKEND=fanout <svc>). config.json
cannot override a baked container env (injection is setdefault).
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.
- 11d ago First seen · 103 lines · 82 tokens per session scan A a37bdc20bb07
epistemic-graph-troubleshooting is a skill published in the GitHub repository Knuckles-Team/epistemic-graph (10 stars, last pushed 13d ago), licensed MIT. It adds 82 tokens to every session and 1,396 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
docker-debug
Debug Kurtosis running on local Docker. Inspect engine, API container, and service logs. Diagnose container crashes, port conflicts, and networking issues. Use when kurtosis commands fail or services aren't reachable on Docker.
k8s-debug-pods
Debug Kurtosis pods on Kubernetes. Diagnose why pods are Pending, CrashLoopBackOff, ImagePullBackOff, or Evicted. Check node taints, tolerations, resource pressure, and pod events. Use when kurtosis engine start fails or pods aren't coming online.
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.
kubernetes
Kubernetes operations playbook for deploying services: core objects, probes, resource sizing, safe rollouts, and fast kubectl debugging.
cluster-health
· Check Kubernetes cluster health and node status with read-only diagnostics after incidents or maintenance.
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…