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 yunsii/wezdeck --skill human-rungit clone --depth 1 https://github.com/yunsii/wezdeckWrote 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/yunsii/wezdeck/human-run)<a href="https://agentmods.dev/skills/yunsii/wezdeck/human-run"><img src="https://agentmods.dev/badge/skills/yunsii/wezdeck/human-run/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/yunsii/wezdeck/human-run"><img src="https://agentmods.dev/badge/skills/yunsii/wezdeck/human-run.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 | $0.00123 | $0.01640 |
| Opus 5.5 | $0.00049 | $0.00656 |
| Sonnet 5.5 | $0.00025 | $0.00328 |
| Haiku 4.5 | $0.00012 | $0.00164 |
Grade A, and why
human-run scanned grade A with 1 finding against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 19d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Makes network callslowCapability
Not a fault in itself. Listed so you know the mod talks to something, and to what.
while ! curl -fsS "$health"; do sleep 5; done How it starts
The opening of the file, as written. The whole thing — 144 lines — stays where its author put it; the contents beside it link to each section on GitHub.
human-run (platform skill — single source)
Who runs the script: the human, via shell command x.
Who prepares it: the agent, via this skill’s ensure-env.sh + propose.sh.
Never paste a multi-line script into chat as the execution channel.
Never wrap agent self-exec through this skill — tools already run those.
Runtime CLI lives in the same wezdeck tree as this skill:
scripts/runtime/cli/{wd-run,x} + agent-run-lib.sh.
Docs: docs/agent-run.md. Profile: agent-profiles → reporting.md.
Who runs what
| Actor | Duty |
|---|---|
| Human | In a real terminal: x (peek → confirm → run in recorded cwd) |
| Any agent | Load skill → ensure-env → propose.sh → tell user only x |
| Agent tools | Self-runnable work — do not call this skill |
When to load (mandatory)
Load whenever any of these is true:
- You are about to ask the human to run / paste / copy a shell script or multi-line command
- The command needs a real TTY, GUI, host secret the agent must not hold, or privileges outside the agent sandbox
- The human must read and approve the exact script before it runs
Do not load when you can run it with tools, or when the snippet is illustrative only (mark 勿粘贴执行).
Resolve TOOL_HOME
1. directory of this SKILL.md (follow symlinks; ~/.agents/skills/human-run → …)
2. $HUMAN_RUN_HOME if set
3. else fail — re-run link-platform-skills.sh from a wezdeck that has human-run
The source checkout may be declared explicitly with `WEZDECK_REPO=/absolute/path`.
Do not use the logical `~/.agents/skills/human-run` path to derive the repo root.
Environment check + init (mandatory first step)
Do not hand-roll path guessing in the chat. Always:
TOOL_HOME="$(readlink -f "${HUMAN_RUN_HOME:-$HOME/.agents/skills/human-run}")"
"$TOOL_HOME/ensure-env.sh"
ensure-env.sh is idempotent. It:
- Verifies this skill’s wezdeck tree has
wd-run/x/agent-run-lib.sh - If
$WEZDECK_REPO/scripts/runtime/clilacks them, symlinks from the skill tree (sowezdeck-envPATH and humanxwork even when that clone’s HEAD is behind) - Writes/updates
~/.wezterm-x/agent-tools.envwith a workingwd_run= - Smokes the binary; prints the absolute
wd-runpath on stdout
What ships with it
3 files beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
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.
- 19d ago Changed 2a4398d74e57
- 20d ago Changed · +3 lines 79c7d5a13560
- 1mo ago First seen · 141 lines · 123 tokens per session scan A e07f028d17b5
human-run is a skill published in the GitHub repository yunsii/wezdeck (5 stars, last pushed yesterday), licensed MIT. It adds 123 tokens to every session and 1,640 once invoked, about $0.0005 per session on Opus 5.5. A static security scan graded it A with 1 finding (makes network calls). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-09-13.
Other skills, from other repositories
agent-self-scheduling
Schedule AI agent runs with cron, loops, or external clocks while avoiding unsafe tight autonomous timers.
agentmail
Use your assigned AgentMail inbox to read email tasks, explicitly send or reply, and check delivery. Provided automatically by your inbox assignment.
feishu
Work with Feishu or Lark bots, docs, sheets, bitables, approval flows, and OpenAPI/MCP setup without hardcoding credentials.
mission-slice-sop
Use when starting, building, handing off, restoring, or closing a mission or slice under the default lightweight SDLC.
material-organizer
Organize only the materials the user has already provided or explicitly pointed to. The job is to make messy inputs easier to scan, reuse, archive, or continue from later while preserving source traceability.
schedule
Manages recurring and one-off scheduled tasks. Session-scoped scheduling via CronCreate/CronDelete/CronList. Documents the cloud path for tasks that need to survive machine sleep or network drops.