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 nxtg-ai/forge-plugin/plugin install nxtg-forgeWrote 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/nxtg-ai/forge-plugin/claude-code-framework)<a href="https://agentmods.dev/skills/nxtg-ai/forge-plugin/claude-code-framework"><img src="https://agentmods.dev/badge/skills/nxtg-ai/forge-plugin/claude-code-framework/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/nxtg-ai/forge-plugin/claude-code-framework"><img src="https://agentmods.dev/badge/skills/nxtg-ai/forge-plugin/claude-code-framework.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 | $0.00138 | $0.04416 |
| Opus 5.5 | $0.00055 | $0.01766 |
| Sonnet 5.5 | $0.00028 | $0.00883 |
| Haiku 4.5 | $0.00014 | $0.00442 |
Grade C, and why
Claude Code Framework scanned grade C 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 1mo 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.
Asks for rootmediumPrivilege escalation
A mod that escalates privileges can change anything on the machine, not only the project.
"deny": ["Read(**/.env*)", "Read(**/*.key)", "Bash(rm *)", "Bash(sudo *)"] Reads agent configuration directoriesmediumAgent snooping
.claude/, .codex/, .gemini/ hold keys, settings and other credentials a mod has no legitimate need for.
4. `~/.claude/settings.json` / `~/.claude.json` — user-level global How it starts
The opening of the file, as written. The whole thing — 382 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Claude Code Framework — Reference & Decision Guide
Authoritative reference for Claude Code's behavior and for authoring its extensions. Ground every claim to current behavior at https://code.claude.com/docs. Model IDs, pricing, and "as of version X" specifics rot — re-verify against live docs before relying on them. This file is the index; depth lives in the linked reference files.
CORE OPERATING RULES
Apply these directly when working with Claude Code on real tasks:
- Plan before code on anything non-trivial. Use Plan Mode (or a written plan) to agree on approach before edits. Discussion-then-code catches wrong turns cheaply.
- Keep CLAUDE.md lean and failure-focused. Document what Claude gets wrong in this repo (build quirks, forbidden paths, naming rules), not what it does right. ~100–200 lines; past ~40k chars Claude Code warns performance degrades — move depth into skills/docs and link them.
- Clear context aggressively.
/clearbetween unrelated tasks; provide only files relevant to the current task. Don't let one session accumulate stale context. - TDD as guardrail. Write/keep failing tests first, let Claude iterate to green. Tests are the objective signal that beats "looks done."
- Split write vs. review contexts. Have a fresh context or subagent verify code the first context wrote — a single context rationalizes its own mistakes.
- Checkpoints ≠ version control.
/rewind(or double-Esc) restores Claude's edits and conversation, but NOT your bash side effects (migrations,rm, pushes). Commit real milestones to git.
EXTENDING CLAUDE CODE — PICK THE RIGHT SURFACE
| You want to… | Use | Lives in |
|---|---|---|
Reusable prompt shortcut (/thing) |
Slash command | .claude/commands/*.md |
| Specialized persona with its own context/tools | Subagent | .claude/agents/*.md |
| Auto-loaded domain knowledge (model-triggered) | Skill | .claude/skills/*/SKILL.md |
| Deterministic action on an event (lint, guard, sync) | Hook | settings.json + script |
| External tool/data integration | MCP server | .mcp.json / claude mcp add |
| Repo-wide standing context | CLAUDE.md | ./CLAUDE.md, ~/.claude/CLAUDE.md |
What ships with it
4 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.
- 1mo ago First seen · 382 lines · 138 tokens per session scan C 8f2619af3136
Claude Code Framework is a skill published in the GitHub repository nxtg-ai/forge-plugin (6 stars, last pushed 4d ago), licensed MIT. It adds 138 tokens to every session and 4,416 once invoked, about $0.0006 per session on Opus 5.5. A static security scan graded it C with 2 findings (asks for root, reads agent configuration directories). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-09-03.
Other skills, from other repositories
rag-retrieval
Retrieval-Augmented Generation patterns for grounded LLM responses. Use when building RAG pipelines, embedding documents, implementing hybrid search, contextual retrieval, HyDE, agentic RAG, multimodal RAG, query decomposition, reranking, or pgvector search.
error-analysis
Evals-first error analysis for LLM apps: clusters real Langfuse or JSONL traces into a human-confirmed failure taxonomy with counts, then recommends binary pass/fail evals for recurring named modes. Use to learn what to measure before writing evals. Not for CI failures.
dev
One-command dev loop boot. Spins up portless (named HTTPS subdomain), emulate (stateful API mocks), the project's dev server, and an agent-browser session, all keyed to the current git branch. Use when starting a feature branch, switching worktrees, or returning to a project after a break. Skips silently with install…
testing-llm
LLM and AI testing patterns — mock responses, evaluation with DeepEval/RAGAS, structured output validation, and agentic test patterns (generator, healer, planner). Use when testing AI features, validating LLM outputs, or building evaluation pipelines.
accessibility
Accessibility patterns for WCAG 2.2 compliance, keyboard focus management, React Aria component patterns, cognitive inclusion, native HTML-first philosophy, and user preference honoring. Use when implementing screen reader support, keyboard navigation, ARIA patterns, focus traps, accessible component libraries…
manage-schemas
Creates, views, edits, deletes, and validates note schemas for the MCP Task Orchestrator in .taskorchestrator/config.yaml — the templates that define which notes agents must fill at each workflow phase. Also recommends schema designs from a library of workflow patterns (autonomous coding loops, spec-driven teams…