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 xcaeser/work-skill --skill work-initgit clone --depth 1 https://github.com/xcaeser/work-skillWrote 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/xcaeser/work-skill/work-init)<a href="https://agentmods.dev/skills/xcaeser/work-skill/work-init"><img src="https://agentmods.dev/badge/skills/xcaeser/work-skill/work-init/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/xcaeser/work-skill/work-init"><img src="https://agentmods.dev/badge/skills/xcaeser/work-skill/work-init.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 | $0.00043 | $0.00554 |
| Opus 5.5 | $0.00017 | $0.00222 |
| Sonnet 5.5 | $0.00009 | $0.00111 |
| Haiku 4.5 | $0.00004 | $0.00055 |
Grade A, and why
work-init 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 1mo 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 — 61 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Work / 1. Init
Create the smallest project foundation that builds and runs. Preserve reliability, user state, simplicity, and clean breaks. Do not add product features, speculative packages, compatibility scaffolding, or infrastructure the foundation does not yet need.
Resolve only essential choices
Inspect the destination and conversation before asking anything. Reuse answers already given. If the request is sufficient, proceed without an interview.
Ask no more than three short questions at a time, in this order:
- What is being built, where should it live, and which platform must it run on?
- Which language, runtime or toolchain, and minimum supported version are required?
- Which packages, services, storage, authentication, and quality commands are genuinely required now?
Offer sensible defaults when local conventions or the ecosystem make them clear. Explain only choices with a meaningful tradeoff.
Before writing, state:
- Foundation goal: one end-to-end capability the scaffold must prove.
- Ready when: build, run, and relevant quality checks succeed.
- Does not count: an empty or unexecuted scaffold.
- Deferred: product features and infrastructure intentionally excluded.
Build in working layers
- Read local instructions and inspect the destination for unrelated files.
- Create an idiomatic scaffold with native metadata and one obvious entry point.
- Add only approved dependencies; prefer platform and existing workspace tooling.
- Format or lint, run focused tests when behavior exists, and perform a clean build.
- Start the project locally when safe and useful.
- Exercise one relevant failure boundary, such as invalid configuration or a missing required dependency.
- Inspect generated files and lockfiles. Report exact commands and evidence.
Do not invent tests for an empty entry point. When the scaffold establishes real behavior, state the behavior being protected, why it matters, the test level, and the exact observable assertion before test edits.
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.
- 1mo ago First seen · 61 lines · 43 tokens per session scan A 1b8316ce83df
work-init is a skill published in the GitHub repository xcaeser/work-skill (2 stars, last pushed 2mo ago), licensed MIT. It adds 43 tokens to every session and 554 once invoked, about $0.0002 per session on Opus 5.5. 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
cortex-init
Set up Cortex by discovering the user's existing systems and generating a personalized Workflow Registry from them. Use when the user runs /cortex-init, says "set up cortex", "adopt my tools into cortex", "build my registry", or right after installing Cortex for the first time.
cortex-learn
Detect routing patterns in the Cortex log and surface candidate Decision Shortcuts for cortex.md. Use when the user asks for routing patterns, wants to tune cortex defaults, or says "/cortex-learn".
cortex-reroute
Mark the last Cortex routing decision as corrected and record the route the user actually wanted. Use when the user says "reroute", "actually use X", "should've been Y", or types "/cortex-reroute".
cortex-log
Show recent Cortex routing decisions with reasoning trail. Use when the user asks what routes have been picked, wants to audit recent routing decisions, or says "show cortex log" / "what did cortex do" / "/cortex-log".
code-review
Orchestrate an adaptive multi-agent review of a pull request or local change that builds a structural and semantic brief, assigns risk-based reviewer lanes at runtime, challenges and deduplicates their findings, verifies each one, and renders an evidence-backed review. Use when a pull request or local change needs…
tech-debt-audit
Audit a whole repository or subsystem for technical debt with at most four non-overlapping read-only subagents, lead verification, and interest-over-principal ranking, and produce an evidence-backed debt register. Use when the user wants to find, size, or prioritize existing debt across a codebase; not for reviewing…