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 3xachris/3xa-harness --skill workordergit clone --depth 1 https://github.com/3xachris/3xa-harnessWrote 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/3xachris/3xa-harness/workorder)<a href="https://agentmods.dev/skills/3xachris/3xa-harness/workorder"><img src="https://agentmods.dev/badge/skills/3xachris/3xa-harness/workorder.svg" alt="Measured on agentmods" 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 | $0.00088 | $0.03124 |
| Opus 5 | $0.00044 | $0.01562 |
| Sonnet 5 | $0.00018 | $0.00625 |
| Haiku 4.5 | $0.00009 | $0.00312 |
Grade A, and why
workorder 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 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.
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 — 113 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Workorder
A frozen order is the contract the build runs against: written before the first edit, unchanged while the work runs, and the only thing acceptance is measured against. It is also the whole brief — what a subagent or a fresh session is handed when the work moves.
Scope stays where it was frozen, so the thing delivered is the thing agreed. And the work runs to done in one pass — obstacles get solved in the same session that hit them, so the human is asked once, at the start, rather than at every bump.
1. Write the order
Seven fields, plus the header line that freezes it.
- One sentence — the whole task, in one. Two sentences means two orders, or a request still in fog: interview it sharp first, write the order after.
- Known facts — what is already verified, each with its source (a prior report, a measurement, a file path). This field is what stops the build re-discovering what was settled last week.
- Materials — the models, datasets, fonts, and dependencies this order may use, each with its licence and the date that licence was checked. Anything the order rules out is listed with the reason, so the check is not repeated either.
- Build spec — every parameter pinned to a file plus a name inside it: a symbol in code, a key in a config, a named preset and the parameter within it for a render, a voice profile and its setting for audio. Whatever the domain, "the value we settled on" has to resolve to exactly one place someone else can open. Where a value genuinely lives only in a GUI, the pin is a written line in the order plus a screenshot in the artifacts folder — a copy someone can check, since the slider cannot be cited. The spec also names that artifacts folder, which holds the order, its gate folders, and its closeout, so the whole task is one place on disk.
- Budget fuse — a number the work counts against, in whatever unit this task actually spends: renders, GPU-hours, API calls, install attempts, drafts, research passes. The count is kept as it goes, in the order's artifacts folder, because a number reconstructed at closeout is a number remembered rather than measured. Reaching the fuse ends the build with a
STOPPEDmessage and returns what exists so far. - Acceptance — objective lines someone else could check without asking what was meant. Each one names how it will be evidenced: a path, a command's output, a measured number. The
DONEcondition is defined byhonest-closeout, including its required verification outside the work's control. - Non-goals — what this order leaves alone, especially the neighbouring work it would be natural to drift into.
What ships with it
1 file 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.
- 7d ago First seen · 113 lines · 88 tokens per session scan A 08d7a3265345
workorder is a skill published in the GitHub repository 3xachris/3xa-harness (17 stars, last pushed 7d ago), licensed MIT. It adds 88 tokens to every session and 3,124 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 2026-08-30.
Other skills, from other repositories
at-review
Review code changes for bugs, regressions, convention violations, and high-value cleanup opportunities. Use for diffs, commit ranges, hosted PR/MR URLs, branches, paths, staged changes, or working-tree changes.
at-vision
Inspect screenshots, photos, diagrams, image paths, and image URLs when the task depends on visible content. Use when the prompt lacks actual image content, native inspection fails, or the user requests inspectimage; prefer the MCP tool, then the installed CLI.
at-usage
Query and display the current API provider balance and recent usage.
delegate
Let Claude Architect route a versioned implementation spec through the trusted MCP runtime, independently review the Candidate Artifact, record a decision, and integrate only accepted bytes. Use for implementation delegation, Producer selection, or commitment-boundary review.
codex
Run the Codex CLI directly in the user's checkout for code analysis, refactoring, or automated editing without Claude Architect's verified delegation lifecycle.
deep-research
Autonomous deep research on codebases and technical topics with structured report output via map-reduce explorer architecture.