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 humanerd-drew/opencode-drewgent --skill huly-selfhost-nas-2026-06-15git clone --depth 1 https://github.com/humanerd-drew/opencode-drewgentWrote 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/humanerd-drew/opencode-drewgent/huly-selfhost-nas-2026-06-15)<a href="https://agentmods.dev/skills/humanerd-drew/opencode-drewgent/huly-selfhost-nas-2026-06-15"><img src="https://agentmods.dev/badge/skills/humanerd-drew/opencode-drewgent/huly-selfhost-nas-2026-06-15/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/humanerd-drew/opencode-drewgent/huly-selfhost-nas-2026-06-15"><img src="https://agentmods.dev/badge/skills/humanerd-drew/opencode-drewgent/huly-selfhost-nas-2026-06-15.svg" alt="Reviewed on agentmods" width="80" 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 5.1 | $0.00050 | $0.02708 |
| Opus 5 | $0.00025 | $0.01354 |
| Sonnet 5 | $0.00010 | $0.00542 |
| Haiku 4.5 | $0.00005 | $0.00271 |
Grade D, and why
huly-selfhost-nas-2026-06-15 scanned grade D 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 7d 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.
- `setup.sh --secret` was chained with `setup.sh --quick` via `&&`: `sudo bash setup.sh --secret && sudo bash setup.sh --quick` Recursive force deletehighDestructive command
rm -rf with a variable or a broad path is one typo away from removing the wrong tree.
- `sudo rm -rf` was blocked with same message How it starts
The opening of the file, as written. The whole thing — 163 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Huly Self-Host on NAS — Session Reference (2026-06-15)
This is a session reference, not a class-level skill. The parent umbrella is software-development/huly-integration. Read that first for the canonical flow; this file is the gotchas and execution notes from a 6+ hour session.
What was actually wrong (the real cause of kvs-1 restart loop)
The kvs-1 container was in restart loop with password authentication failed for user huly. After 5+ hours of investigation, the actual root cause was:
setup.sh --secretwas chained withsetup.sh --quickvia&&:sudo bash setup.sh --secret && sudo bash setup.sh --quick- The first call worked: wrote
.huly.secret,.cr.secret,.rp.secret(65-byte hex each) - The second call's
envsubststep failed silently because the NAS doesn't haveenvsubst(nogettextpackage), AND the second invocation's logic was confused by the first call's outputs - Result:
huly_v7.confended up 0 bytes (empty file) setup.sh --quickcontinued, wrote nginx.conf, randocker compose up -d- Compose read empty
.env→ all variables blank → allWARN[0000] The "X" variable is not set. Defaulting to a blank string.warnings hulyuser was never created in cockroachdocker compose up -dcompleted but kvs-1, account-1, workspace-1, etc. all failed password auth againsthulyuser (which didn't exist)- Front-1 returned HTTP 200 (it just serves static) → masking the actual failure
Lesson: If you see kvs-1 in restart loop with password authentication failed for user huly, check wc -l huly_v7.conf first. If it's 0, the envsubst failed and you need to hand-fill the file. Skip everything else.
The credential-masking trap with expect
The security system masks anything that looks like a password with ***. When you use expect to send a script that contains secret values, the *** literal can leak into the command and break things.
What went wrong:
# This looked fine in the expect source but failed when sent over SSH
send "sed -i 's|^CR_USER_PASSWORD=.*|CR_..._V|' huly_v7.conf\r"
# Expect's Tcl parser interpreted $CR_PW or similar as a variable
# When it sent, the actual literal value was `*** ` placeholder text
# NAS file ended up with `CR_USER_PASSWORD=*** (literal 3-star)
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.
- 7d ago First seen · 163 lines · 0 tokens per session scan D 89ed0cb76ffa
huly-selfhost-nas-2026-06-15 is a skill published in the GitHub repository humanerd-drew/opencode-drewgent (2 stars, last pushed 1mo ago), licensed MIT. It adds 50 tokens to every session and 2,708 once invoked, about $0.0003 per session on Opus 5. A static security scan graded it D with 2 findings (asks for root, recursive force delete). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-09-03.
Other skills, from other repositories
ciel-project-scaffolder
CIEL's framework for standardized project initialization, module creation, and boilerplate management.
shotgrid-setup
ShotGrid setup skill - generate local IDE, mcpcall, Docker, gateway, and custom skill injection configuration for dcc-mcp-fpt. Use when an agent needs to help a user bootstrap or audit a local/container deployment.
scaffold
Bootstrap a new project with your stack, auth, database, and standards pre-configured.
deploy-docker-compose
Run the Omnigent server as a Docker compose stack (server + Postgres) on any Docker host — your laptop, a VPS, EC2 by hand, or as the base layer of any container-platform deploy. Invoke when the user wants to build the image, bring up the compose stack, debug the stack on a host they already have, or extend the stack…
workthreads
SpecStory Workthreads - a weekly work-thread rollup across a team's repos from SpecStory coding histories (any agent - Claude Code, Codex, Cursor, Gemini, and more). It groups the window's sessions into threads of work per project and labels each new / open / recently closed, so a lead sees what shipped, what is still…
atmos-config
Atmos root configuration: atmos.yaml discovery, precedence, deep merging, basepath, imports, minimal bootstrap, and routing to narrower Atmos skills.