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/syntax753/synkb/launch-demonpx skills add Syntax753/synkb --skill launch-demogit clone --depth 1 https://github.com/Syntax753/synkbWhat 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 | $0.00058 | $0.02176 |
| Opus 5 | $0.00029 | $0.01088 |
| Sonnet 5 | $0.00012 | $0.00435 |
| Haiku 4.5 | $0.00006 | $0.00218 |
Grade C, and why
launch-demo 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 2d 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.
Downloads and executes remote codehighSupply chain
curl | sh runs whatever the server returns today, which is not necessarily what it returned when this was reviewed.
curl -fsS http://localhost:8086/library/items | python -c "import sys,json;print(len(json.load(sys.stdin)['items']))" Makes network callslowCapability
Not a fault in itself. Listed so you know the mod talks to something, and to what.
curl -fsS http://localhost:8080/healthz # -> {"status":"ok","service":"core-api"} How it starts
The opening of the file, as written. The whole thing — 160 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Launch the KB Platform demo
The demo is a Docker Compose stack (docker-compose.demo.yaml, 16 services) fronted by a
Next.js web app on http://localhost:3000. Everything is driven through the Makefile.
Never call docker compose without -f docker-compose.demo.yaml — the compose file is
non-default, so a bare docker compose finds no services.
One command
make demo
make demo is the idempotent "make me demo-ready" target. It:
- pre-cleans stopped containers from previous runs,
- brings the stack up if it's down (
docker compose up -d --build, rebuilds only changed images), - (re-)seeds the demo data — 50 docs, 8 templates, 5 drafts, ~80 audit events,
- prints the URL and what to try.
Safe to re-run between demos without losing state. On success it prints:
Web: http://localhost:3000 (logged in as demo user — all groups)
Script: demo/demo-script.md (14-step printable cue card)
Build time: a cold image cache (first-ever run, or after make reset) can take anywhere
from a few minutes to ~an hour — the mcp-server / torch-pulling services dominate. A warm
stack finishes in well under a minute. On Windows, prefer plain build output for backgrounded
builds: docker compose -f docker-compose.demo.yaml build --progress=plain (BuildKit TTY
output is lossy when redirected).
Sub-option: run on your personal Claude account — no API key (SUPPLY_PROXY=1, WP-31)
To make the Chatbot + draft generation work without an ANTHROPIC_API_KEY, route the
demo's LLM calls through your own Claude subscription via the opt-in claude-proxy:
make demo SUPPLY_PROXY=1
It reads your Claude OAuth token once on the host (~/.claude/.credentials.json; never mounted
into a container), starts an in-network proxy (claude-proxy:3456) that forwards to
api.anthropic.com under that token, and fails the demo loudly if the token is rejected
(with a "validate your token … or set ANTHROPIC_API_KEY" message). A real ANTHROPIC_API_KEY
(env or .env) always wins and skips the proxy.
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.
- 2d ago First seen · 160 lines · 58 tokens per session scan C c93fc8cca51d
launch-demo is a skill published in the GitHub repository Syntax753/synkb (0 stars, last pushed 1mo ago), licensed Apache-2.0. It adds 58 tokens to every session and 2,176 once invoked, about $0.0003 per session on Opus 5. A static security scan graded it C with 2 findings (downloads and executes remote code, makes network calls). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.
Other skills, from other repositories
agent-authority-charter-builder-arkadiy-miteiko
Creates an Agent Authority Charter for enterprise or regulated AI agents before deployment. Use this Skill when a user needs to define what an AI agent is allowed to do, who delegated authority to it, what actions are permitted or prohibited, when human approval is required, what evidence must be preserved, and how…
vps-docker
Full VPS Docker status — all containers, resource usage, arifOS health.
craig-voice
The Craig Horton voice, and what he hates. Load BEFORE writing anything in Craig's name or the AI BVF brand - Transformation Brief essays, LinkedIn or X posts, site and README copy, engine output text, PR descriptions, marketing copy, branded PDFs and decks. Also load when asked to review a draft ("what i hate", "is…
aibvf-design
Use this skill to generate well-branded interfaces and assets for AI BVF (the AI Business Value Framework, by Craig Horton Advisory), either for production or throwaway prototypes/mocks/etc. Contains essential design guidelines, colors, type, fonts, assets, and UI kit components for prototyping.
aiq-deploy
Use when asked to install, deploy, run, validate, troubleshoot, or stop NVIDIA AI-Q Blueprint infrastructure.
cuopt-install
Install cuOpt for Python, C, or server via pip, conda, or Docker; verify the install. For building cuOpt from source, see cuopt-developer.