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 jerseycheese/agent-skills --skill backlog-routinggit clone --depth 1 https://github.com/jerseycheese/agent-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/jerseycheese/agent-skills/backlog-routing)<a href="https://agentmods.dev/skills/jerseycheese/agent-skills/backlog-routing"><img src="https://agentmods.dev/badge/skills/jerseycheese/agent-skills/backlog-routing/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/jerseycheese/agent-skills/backlog-routing"><img src="https://agentmods.dev/badge/skills/jerseycheese/agent-skills/backlog-routing.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 | $0.00075 | $0.04717 |
| Opus 5.5 | $0.00030 | $0.01887 |
| Sonnet 5.5 | $0.00015 | $0.00943 |
| Haiku 4.5 | $0.00007 | $0.00472 |
Grade A, and why
backlog-routing 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 — 350 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Backlog routing
The job is throughput without losing control. Lots of agents, each on the cheapest model that can actually do its piece, none of them stepping on each other, and nothing merged unless the repo's merge policy allows it. By default that means the human reads every PR first.
It's built from skills that already exist. This one decides what runs where, in what order, on which model. The rest is delegated:
| Need | Delegate to |
|---|---|
| Which provider and model | route (your routing table plus current burn) |
| Ranking | prioritize-issues |
| Labels, dedupe, clustering | issue-maintenance |
| Visual finding triage | visual-qa-pipeline (its severity ladder) |
| Work inside a lane | analyze-issue → tdd-implement → ship-issue steps 1–5 |
| CI red | ci-fix-with-memory |
| After merge | post-merge |
| Claims of done | evidence-check |
Don't re-implement any of those here.
Requirements. GitHub for issues, milestones, sub-issues and PRs (gh, or a GitHub MCP
server). Any agent host works for the orchestrator. Where this skill names a specific tool (Claude
Code, Codex, Antigravity), it's an example of a kind of lane, not a requirement.
0. The adapter comes first
Every repo that uses this needs an adapter skill (for example <repo>-backlog-routing). Read it
before any stage. If there's no adapter, stop and write one first, using the contract below. Don't
guess a repo's gate commands or release process.
The adapter supplies:
- Base branch and any branch that's off limits (a release-only
main, say). - Gate commands a lane must show passing, and which extra gates trigger on which kinds of change.
- Local-only criteria: what can't be proven in a cloud container (OS-specific visual baselines, live API keys that live in the user's browser, a human walking a flow, headful playtests).
- Merge policy: who merges, and what must be shown first. With no policy the default is human-only: no agent merges, approves, or enables auto-merge. An adapter can give merging to the orchestrator for the PRs it names (say, everything except drafts and PRs with a Local check), and lists the conditions to show first (gates, CI, a clean review gate). It also says how long to wait for an automated reviewer.
- Milestone and release convention: how milestones are named, what a release PR contains, which release steps belong to the human.
- Label names for type, priority, size, model tier, and the two this skill needs (
needs-local,run-tracker). Use only labels that exist in the repo. - PR template path and any rule about rendering it.
- Trap list: failure modes lanes have actually hit in this repo. It goes into every brief.
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.
- today Changed · -9 lines · -171 tokens per session 0ef8b4032e25
- 3d ago Changed · +27 lines · +10 tokens per session 62ee7b0ef749
- 19d ago First seen · 332 lines · 236 tokens per session scan A 986019247a6f
backlog-routing is a skill published in the GitHub repository jerseycheese/agent-skills (3 stars, last pushed today), licensed MIT. It adds 75 tokens to every session and 4,717 once invoked, about $0.0003 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 2026-09-27.
Other skills, from other repositories
omh-github-issue-intake
This is an OMH github-issue-intake workflow skill, projected for Agent Skills hosts (Claude Code, Codex, Cursor, opencode, OpenClaw, pi).
nav-pilot
Dispatch a Navigator task doc to Pilot as a labeled GitHub issue. Use when user says "dispatch TASK-XX to Pilot", "hand off to Pilot", "send to Pilot", or "queue for Pilot". One-way handoff (Navigator authors the spec, Pilot executes autonomously).
lov-rename-project
A project-renaming workflow that updates a product or repository name across source files, configuration, and documentation. It produces a reviewable plan and preserves compatibility-sensitive names and references where needed.
lov-gh-tidy
Interactive GitHub repo hygiene skill. Lists all open issues, PRs, stale branches, and orphan labels, shows a summary of each with analysis, then asks the user how to handle each item (close, merge, comment, delete, keep). Executes all chosen actions via gh CLI. Use when the user says "清理 GitHub", "tidy repo", "clean…
github-project-management
GitHub project management with issue tracking, project-board automation, and sprint planning. Use when triaging or organizing issues, automating a project board, planning a sprint, or running release/roadmap coordination on GitHub.
morning-brief
Operator's read-only morning view of the current GitHub repo in one pass: open counts per queue label, merge-ready PRs, parked decisions with their RECOMMENDED lines, and loop-lane telemetry freshness; never mutates issues, PRs, labels, or comments. Use when: 'morning brief', 'ops dashboard', 'what needs attention'…