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 asfbay-bit/opchain-skills --skill ow-startgit clone --depth 1 https://github.com/asfbay-bit/opchain-skillsWrote 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/asfbay-bit/opchain-skills/ow-start)<a href="https://agentmods.dev/skills/asfbay-bit/opchain-skills/ow-start"><img src="https://agentmods.dev/badge/skills/asfbay-bit/opchain-skills/ow-start/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/asfbay-bit/opchain-skills/ow-start"><img src="https://agentmods.dev/badge/skills/asfbay-bit/opchain-skills/ow-start.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 | $0.00223 | $0.03442 |
| Opus 5.5 | $0.00089 | $0.01377 |
| Sonnet 5.5 | $0.00045 | $0.00688 |
| Haiku 4.5 | $0.00022 | $0.00344 |
Grade A, and why
ow-start 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 today.
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 — 151 lines — stays where its author put it; the contents beside it link to each section on GitHub.
ow-start: the fit call, its recap, and the logs
ow-start 2.0.1 · protocol 2 · Core bundle
Open references/ow-protocol.md first and keep to it. If you cannot open it, say so and stop.
Asking the user: every question and approval goes through AskUserQuestion, batched into one call per verb with your recommendation first; never a question or numbered menu typed into chat (references/ow-protocol.md §1).
ow-start prepares the 20-minute call, then writes the recap the client receives: Today · What needs to change · Next step · What I need from you, one recommended path, due within one business day. It also keeps the logs: the meetings no other ow- skill records, and the decisions, per client and for your business.
Read first. Every verb starts from the latest saved version of what it depends on in the CRM (below), never its own earlier file; prices, numbering, client profile and terms come from there. Where the CRM cannot be read or a route is not built, use the scratch cache in opchain-work/ (profile.md, offers.md, handoffs.yaml) and say so in the output. Never claim a publish the call did not confirm.
Go with the flow: the user may skip a step, redo one, or jump straight to the recap from a transcript. Ask only what no file answers, state every assumption in one line the user can correct, and keep moving.
Welcome (first run: nothing in the CRM or the cache)
At most four lines, then examples:
I prepare your 20-minute fit calls and write the client recap afterwards. Before the call: a one-page call guide and a few questions to send ahead. After it: give me the transcript or your notes, and I write the recap with one recommended path. Your CRM holds the record;
opchain-work/in the folder you attach is the scratch copy.
Try: "prep my call with Acme on Thursday" · "here's the transcript, write the recap" · "log my check-in with Acme" · "where did I leave off?"
Verbs
| Verb | When | Does | Writes (under opchain-work/) |
|---|---|---|---|
/ow-start |
Any time | First run: the welcome. Later: resume from the CRM and the cache: open intakes and open logs items | nothing |
/ow-start-prep |
Before the call | One AskUserQuestion call, then the call guide: each bank question marked already answered (with source), must-ask or if-time, the must-asks fitted into 20 minutes |
intake/<client>/intake.yaml, call-guide.md, send-ahead.md |
/ow-start-extract |
After the call | From a transcript or notes: the intake record, coverage, then the client recap with one recommended path. Seeds the profile on first run | intake/<client>/recap.md, intake-record.md, coverage.md; profile.md (first run only) |
/ow-setup |
Any time | Edits the profile; every question can be skipped. Creates offers.md and handoffs.yaml from the examples when they are missing |
profile.md; offers.md and handoffs.yaml (when missing; then they are yours) |
/ow-status |
Any time | One screen from the CRM and the cache: open intakes, open logs items, what is awaited and by when | nothing |
/ow-start-transcribe |
After the call, with a recording | Provided by the ow-tools add-on (references/phase-setup.md §3) |
intake/<client>/transcript.md, by the add-on |
/ow-start-prep-meeting <workstream> |
Before a meeting | An agenda of the open items, each with its source; asks purpose, date, length and attendees in one AskUserQuestion call |
MTG-nn/agenda.md |
/ow-start-log-meeting <workstream> |
After it (MTG-nn optional) |
Notes from a transcript or your notes, every item cited; a follow-up draft when an external party attended | MTG-nn/notes.md, follow-up.{html,txt}; the held line |
/ow-start-action <workstream> <id> <state> |
Any time | A-nn done, A-nn dropped, A-nn moved <date>, DEC-nn reviewed |
one STATUS line |
/ow-start-log-decision <workstream> |
Any time, or from a meeting's notes | One decision per file; a change is a new decision that supersedes it | decisions/DEC-nn.md; a superseded review's closing line |
/ow-start-log <workstream> |
Any time | One screen of meetings, actions, decisions and reviews, with other skills' records by path | nothing |
What ships with it
18 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.
- CHANGELOG.md 11 KB
- examples/handoffs.yaml 4.2 KB
- examples/offers.md 5.8 KB
- references/email-template.md 4.9 KB
- references/intake-record-schema.md 6.1 KB
- references/logs-schema.md 7.8 KB
- references/logs-status-and-index.md 6.6 KB
- references/ow-protocol.md 5.9 KB
- references/phase-extract.md 10 KB
- references/phase-log-decision.md 3.0 KB
- references/phase-log-meeting.md 4.4 KB
- references/phase-prep-meeting.md 2.7 KB
- references/phase-prep.md 6.1 KB
- references/phase-setup.md 1.6 KB
- references/question-bank-delivery.md 2.4 KB
- references/question-bank-prospect.md 3.1 KB
- references/self-checks-logs.md 2.3 KB
- skill.edges.yaml 12 KB
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.
- today Changed · -45 lines b0c8d3dafe63
- 5d ago First seen · 196 lines · 223 tokens per session scan A 3fbe4a64eb71
ow-start is a skill published in the GitHub repository asfbay-bit/opchain-skills (0 stars, last pushed today), licensed Apache-2.0. It adds 223 tokens to every session and 3,442 once invoked, about $0.0009 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 2026-10-30.
Other skills, from other repositories
local-ai-agents
Build local-first AI agents that run entirely on a developer workstation with Microsoft Foundry Local and Qwen function-calling models. Covers Small Language Models (SLMs), the OpenAI-compatible local endpoint, sandboxed local tools, local RAG with Chroma, local MCP servers, hybrid cloud/local routing, and the…
systematic-debugging
Use when encountering any bug, test failure, or unexpected behavior, before proposing fixes.
next-cache-components-adoption
Turn on Cache Components in a Next.js app and resolve the blocking routes it surfaces. Use when the user wants to enable, adopt, or migrate to Cache Components, flip the cacheComponents flag, work through a flood of blocking-prerender / instant validation errors, run the cache-components-instant-false codemod, or…
chat-pet-sprite-creation
Use when creating or changing VS Code chat pet sprite art, sprite sheets, state animations, eye treatments, Stable/Insiders variants, or pet transitions under src/vs/workbench/contrib/chat/browser/widget/media/chatPet.
cpu-profile-analysis
Analyze V8/Chrome CPU profiles (.cpuprofile) and DevTools trace files (Trace-.json). Use when: profiling performance, investigating slow functions, comparing code paths, finding bottlenecks, analyzing timeToRequest, understanding call trees from sampling profiler data, analyzing layout/paint/rendering, investigating…
insight-error-page
Write or audit an insight-kind error page for the Next.js dev overlay. Use when creating a new errors/ .mdx page, auditing an existing one, or checking that a page matches the framework fix cards. Covers page structure, title alignment, FixCard cards with Copy prompt button, code snippets, terminology verification…