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 okqixiaobao727-design/agentcrew-dev-skills --skill routegit clone --depth 1 https://github.com/okqixiaobao727-design/agentcrew-dev-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/okqixiaobao727-design/agentcrew-dev-skills/route)<a href="https://agentmods.dev/skills/okqixiaobao727-design/agentcrew-dev-skills/route"><img src="https://agentmods.dev/badge/skills/okqixiaobao727-design/agentcrew-dev-skills/route/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/okqixiaobao727-design/agentcrew-dev-skills/route"><img src="https://agentmods.dev/badge/skills/okqixiaobao727-design/agentcrew-dev-skills/route.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 | $0.00052 | $0.01431 |
| Opus 5 | $0.00026 | $0.00715 |
| Sonnet 5 | $0.00010 | $0.00286 |
| Haiku 4.5 | $0.00005 | $0.00143 |
Grade A, and why
route 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 11d 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 — 106 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Route
/route classifies every ticket of a run on four dimensions — workflow, executor, model, effort —
plus the review lane the reviewed workflows carry, writes each conclusion into its ticket as a
## Routing section, and ends by staging the run directory whose exact /crew command it prints.
That section is advisory input: /crew's driver builds and validates its wave table from it,
and from that validation onward the wave table is the sole routing authority.
Three entrances, one exit — whichever was typed, the run ends at the staging script:
| Typed | The run's ticket set |
|---|---|
/route #<parent> |
the parent ticket's open sub-issues |
/route #<a> #<b> … |
exactly those tickets, whether or not they share a parent |
/route <spec-or-feature-dir> |
cut from that spec first — to-tickets+route |
Routing is a suggestion: steps 1–3 read and propose, step 4 is the first write, and the user decides at the one checkpoint between them.
1. Settle the mode
A repo with no agentcrew.toml at its root is a first run: go to the setup wizard in
references/setup.md, and resume here once it is done. The same wizard
reconfigures a repo whenever the user asks, and fills a config an upgrade left a key short.
Settle the tracker: references/trackers.md turns this repo's
docs/agents/issue-tracker.md into the read, edit, and mark operations the steps below
call.
Then settle the entrance from what was typed, per the table above. A spec whose tickets are not cut
yet is to-tickets+route: hand over to
references/to-tickets+route.md — /mattpocock-skills:to-tickets cuts,
confirms, publishes and stages there with the routing rules riding along, and steps 2–4 below never
run. The other two entrances are route-only: assemble the ticket set — a parent's open
sub-issues, or the tickets named — and read every ticket at the tracker: on github, its body
and every comment; on local, its ticket file. Include Blocked by: edges resolved into dependents
as well as blockers, and the existing ## Routing section. Read the parent at the tracker too
where the set came from one; a set given as a list may have no parent and no spec, and then each
ticket's tracker content is the whole context.
What ships with it
6 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.
- 11d ago First seen · 106 lines · 52 tokens per session scan A 9e7fe6830be5
route is a skill published in the GitHub repository okqixiaobao727-design/agentcrew-dev-skills (1 stars, last pushed 3d ago), licensed MIT. It adds 52 tokens to every session and 1,431 once invoked, about $0.0003 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 2026-08-31.
Other skills, from other repositories
gstack-sprint
3-Phase Sprint workflow — design → execute → review with user interaction at decision points.
review-switch
Dispatch a review to the Bridge, which resolves the reviewer Lane from this machine's Machine Config. Use when a review is asked for without naming a lane, or when a review-skill invocation was refused and pointed here.
task-decomposition
A tool for breaking an approved technical specification into small, separate task files. A technical specification is a document describing what software should do.
feature-execution
A feature-delivery coordinator that divides a feature into waves of work, assigns agents, manages review rounds, and commits each wave.
project-planning
A project-planning process that asks questions, records technical decisions, and fills in the project's documentation. It is designed to complete the project knowledge in one session.
progress
A progress reporter that reads project files and creates a short summary in Russian. It uses git history, checkpoint.yml, task files, and decisions.md, then sends the full progress record to a dashboard API.