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 agentmods add skills/bostonorange/claude-code-framework/setupnpx skills add BostonOrange/claude-code-framework --skill setupgit clone --depth 1 https://github.com/BostonOrange/claude-code-frameworkWhat 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 | $0.00043 | $0.02635 |
| Opus 5 | $0.00022 | $0.01318 |
| Sonnet 5 | $0.00009 | $0.00527 |
| Haiku 4.5 | $0.00004 | $0.00264 |
Grade B, and why
setup scanned grade B with 2 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 yesterday.
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.
Reads agent configuration directoriesmediumAgent snooping
.claude/, .codex/, .gemini/ hold keys, settings and other credentials a mod has no legitimate need for.
grep -r '{{' CLAUDE.md .claude/ 2>/dev/null | grep -vE '\.claude/(state|backup)/' | head -20 Makes network callslowCapability
Not a fault in itself. Listed so you know the mod talks to something, and to what.
Both the detector and applier are *instructed* not to make network calls — no `npm view`, no `gh api`, no `curl`/`wget`/registry queries. The framework's `guardrails.sh` does not enforce this at the harness level (it blo How it starts
The opening of the file, as written. The whole thing — 180 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Setup — First-Time Onboarding
setup.sh puts the framework files on disk. /setup makes them fit your project.
The bash installer can only ask the eight questions it was built around. /setup walks 17 layers — language, framework, build, test, type-check, format/lint, persistence, API style, frontend, design system, monorepo, observability, infra, CI/CD, tracker, notification, branch — and uses what's actually in the repo to pre-fill answers. For each layer it can't auto-answer, it shows the options, the tradeoffs, and a recommended default. You pick "default" or pick differently — no research required.
Lifecycle vs /improve: /setup decides shape (first run, or re-baseline). /improve keeps shape in tune as the project grows. They don't overlap — framework-improver-detector reads the ## Layers owned by /setup block in setup-applied.md to build a skip-list at proposal time, and framework-improver-applier re-validates that skip-list at apply time, halting if anything escaped detection.
Usage
/setup full onboarding pass; detect → propose → confirm → apply
/setup --refresh re-run all 17 layers using existing values as the baseline
/setup --layer=<name> re-run a single layer (e.g., /setup --layer=ci-cd)
/setup --dry-run detect + propose only; do not apply even after confirmation
Layer names: language, framework, build, test, type-check, format-lint, persistence, api-style, frontend, design-system, monorepo, observability, infra, ci-cd, tracker, notification, branch.
Process
Phase 1: Detect (read-only)
Spawn project-setup-detector:
Run your full process. Inventory the repo, classify greenfield vs brownfield,
walk the 17 layers, and write the proposal to .claude/state/setup-proposal.md
following the schema in your Phase 4. Do not apply changes — you don't have
Edit/Write tools. End with the surface summary.
The detector is read-only by tool restriction (no Edit/Write). It writes only .claude/state/setup-proposal.md (creating .claude/state/ via mkdir -p if missing).
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.
- yesterday First seen · 180 lines · 43 tokens per session scan B 72857a7de138
setup is a skill published in the GitHub repository BostonOrange/claude-code-framework (2 stars, last pushed 1mo ago), licensed MIT. It adds 43 tokens to every session and 2,635 once invoked, about $0.0002 per session on Opus 5. A static security scan graded it B with 2 findings (reads agent configuration directories, makes network calls). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.
Other skills, from other repositories
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…
babysit-pr
Babysit a GitHub pull request after creation by continuously polling review comments, CI checks/workflow runs, and mergeability state until the PR is merged/closed or user help is required. Diagnose failures, retry likely flaky failures up to 3 times, auto-fix/push branch-related issues when appropriate, and keep…
imagegen
Generate or edit raster images when the task benefits from AI-created bitmap visuals such as photos, illustrations, textures, sprites, mockups, or transparent-background cutouts. Use when Codex should create a brand-new image, transform an existing image, or derive visual variants from references, and the output…
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…
next-cache-components-optimizer
Drive a Next.js route to instant navigation by setting up an agentic loop, under Cache Components / PPR, on initial load (hard navigation) and client-side navigation (soft navigation). Encode the goal as a failing @next/playwright instant() e2e and work it to green, one verified route at a time; the shipped test then…