Getting it into your agent
This one installs as part of its plugin. Adding the marketplace and installing the plugin brings it with everything else the plugin ships.
/plugin marketplace add wayne930242/straw-boss/plugin install straw-bossWrote 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/wayne930242/straw-boss/choosing-graph)<a href="https://agentmods.dev/skills/wayne930242/straw-boss/choosing-graph"><img src="https://agentmods.dev/badge/skills/wayne930242/straw-boss/choosing-graph.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 | $0.00079 | $0.00979 |
| Opus 5 | $0.00039 | $0.00490 |
| Sonnet 5 | $0.00016 | $0.00196 |
| Haiku 4.5 | $0.00008 | $0.00098 |
Grade A, and why
choosing-graph 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 5d 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 — 44 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Overview
Two choices, made before the first turn of real work and stated out loud: the coordination graph — how the agents on this task are wired — and the reality anchor — the contact with reality that proves the result. Both are coordination, so a main agent fixes both when it dispatches and the anchor travels with the brief. A dispatched worker states its own graph for its own task and works inside the anchor it was handed.
Naming the anchor is not naming the tests. The anchor fixes the category and its checkpoint; inside it the worker and the user choose the method — which seam, which cases, which tool. That split is docs/roles.md's, and it keeps the brief clear of the worker's own work definition.
Both are stated, not asked. The human anchor carries the one user question described below.
Coordination graphs
Which graph applies follows from how the work actually runs.
- single-loop — one agent carries uncomplicated work whose length is visible end to end. A coordinator driving one dispatch's lifecycle, or a worker using one coworker as a check, remains a single loop.
- sub-agent fan-out/fan-in — a main agent or dispatched worker sends clear, converged branches to subagents and integrates their results.
- orchestrator-worker — the coordinator runs multiple app-rooted workers through status events. A dependency graph and a capped batch use this shape. It is the only graph that writes
~/.straw-boss/plans/<slug>/plan.json; the other two carry no dispatch plan.
orchestrator-worker is settled ahead of the other two: more than one app-rooted worker under one coordination loop keeps that shape whatever else runs beside it.
single-loop and sub-agent fan-out/fan-in create no plan.json and no repo-internal Straw Boss planning or spec document. An app-rooted dispatch still writes its own ~/.straw-boss/dispatch/<app>--<slug>.json instruction and .contract.md; these are the dispatch's lifecycle record, archived once the dispatch wraps up.
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.
- 5d ago First seen · 44 lines · 79 tokens per session scan A 658cf6ebca55
choosing-graph is a skill published in the GitHub repository wayne930242/straw-boss (3 stars, last pushed yesterday), licensed MIT. It adds 79 tokens to every session and 979 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 2026-08-31.
Other skills, from other repositories
push-ci
Push to remote and monitor CI. Validates branch safety, executes git push WITH explicit user approval, then monitors CI run status via gh CLI. Use when: user says 'push', 'push and watch CI', 'ship it', 'push-ci'. Not for: committing (use /smart-commit), creating PRs (use /create-pr), merging (use /merge-prep).
create-request
Create, update, or scan per-task request tickets for progress tracking. These are date-prefixed non-lifecycle docs under requests/, NOT feature-level requirements (use /req-analyze for those). Use when: tracking task progress, updating completion status, scanning incomplete requests, checking request status dashboard.…
smart-rebase
Smart partial rebase for squash-merge repositories. Auto-detect which commits to keep/drop when base branch was squash-merged into target. Use when: user says 'rebase', 'partial rebase', 'base already merged', 'smart rebase', or /smart-rebase. Not for: simple git rebase (the developer runs it — Claude never executes…
remind
Lightweight model correction with context-aware rule loading. Use when: model forgot a rule, skipped a required step, edited code/docs without running review, needs to re-read CLAUDE.md or rules. Triggers on: 'you forgot', 'remind', 'check rules', 'what did you miss', '你忘了', 'did you skip review', 'why didn't you run…
adr
Write an Architecture Decision Record (ADR) for a feature — Context / Decision / Status / Consequences / Alternatives, filed as docs/features/ /adr- - .md with a 3-digit zero-padded number. Handles the Superseded case: bidirectional linking when a new ADR replaces an old one. Use when: recording why an architectural…
claude-health
Claude Code config health check + plugin sync. Use when: auditing .claude/ structure, checking naming, verifying hook setup, detecting plugin version drift, syncing installed assets. Not for: skill quality (use skill-health-check), code review (use codex-code-review). Output: health report + fix recommendations.