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 instructions/saitarrun/devforge-ai/claude-mdgit clone --depth 1 https://github.com/saitarrun/Devforge-aiWrote 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/instructions/saitarrun/devforge-ai/claude-md)<a href="https://agentmods.dev/instructions/saitarrun/devforge-ai/claude-md"><img src="https://agentmods.dev/badge/instructions/saitarrun/devforge-ai/claude-md.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 | $0.01274 | $0.01274 |
| Opus 5 | $0.00637 | $0.00637 |
| Sonnet 5 | $0.00255 | $0.00255 |
| Haiku 4.5 | $0.00127 | $0.00127 |
Grade B, and why
Devforge-ai CLAUDE.md scanned grade B with 1 finding 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 3d 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.
Reads agent configuration directoriesmediumAgent snooping
.claude/, .codex/, .gemini/ hold keys, settings and other credentials a mod has no legitimate need for.
- **Permission prompts**: Add patterns to `~/.claude/settings.local.json` under `permissions.allow` How it starts
The opening of the file, as written. The whole thing — 121 lines — stays where its author put it; the contents beside it link to each section on GitHub.
SDLC Workflow Plugin
A multi-agent SDLC orchestration system: 10 agents × 5 phases → plan, build, verify, ship, operate. Tracer bullet development with Ralph Loop self-correction and Linear issue integration.
Project
- Type: Claude Code plugin
- Version: 2.0.0
- Repo: github.com/saitarrun/devforge-ai
Repository Map
agents/— 10 role-specific agents (product-manager, ux-designer, fullstack-engineer, qa-engineer, security-engineer, performance-engineer, devops-engineer, sre-engineer, data-engineer, technical-writer)skills/— 34 knowledge skills (architecture, testing, code-review, ralph-loop, grill-me, etc.)commands/— 10 orchestrator commands (/sdlc, /sdlc-plan, /sdlc-build, /sdlc-verify, /sdlc-ship, /sdlc-operate, /sdlc-review, /sdlc-implement, /to-prd, /to-issues)scripts/— CLI, install, uninstall, validation helpersREADME.md— User documentation and feature overview
Commands
- Install:
npm run install-localormake install-local - Validate:
make validate - Uninstall:
make uninstall
Agent Conventions
Frontmatter required: name, description, tools (comma-separated), model (haiku/sonnet/opus)
Model tiers: Haiku (templated), Sonnet (development), Opus (architecture/security decisions)
Tools: Include only what the agent uses — Read, Write, Edit, Bash, Glob, Grep, WebFetch
- Note: When querying codebase, prefer code-review-graph (bundled integration) over grep for structural analysis. See "Codebase Querying Strategy" below.
Embed book principles (QUANTS/INVEST/Testing Pyramid, Architecture: ADR/fitness functions, Pragmatic Programmer DRY/ETC, Clean Code naming/SOLID) in agent bodies.
Skill Conventions
Structure: skills/skill-name/SKILL.md (no tools, no model — pure knowledge)
Frontmatter: name, description, version
Content: Methodology, checklists, reference patterns. Load on-demand, not always-on.
Agent-Skill Auto-Wiring
All 10 agents are auto-wired to phase-specific skills. When an agent is invoked:
- Agent frontmatter declares
skills: [skill-a, skill-b, ...] - Command loads all phase skills into agent context (pure methodology, not tools)
- Agent system prompt says: "You have access to these skills: X, Y, Z. Apply them."
- Agent applies skill principles → polished industry-standard outputs
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.
- 3d ago First seen · 121 lines · 1,274 tokens per session scan B c34ffca2344b
Devforge-ai CLAUDE.md is an instructions file published in the GitHub repository saitarrun/Devforge-ai (5 stars, last pushed 20d ago), licensed Apache-2.0. It adds 1,274 tokens to every session, about $0.0064 per session on Opus 5. A static security scan graded it B with 1 finding (reads agent configuration directories). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.
Other instructions, from other repositories
Sdlc-ai-workflow CLAUDE.md
Claude Code instructions for saitarrun/Sdlc-ai-workflow, covering sdlc workflow plugin, project, repository map, commands and agent conventions.
memi CLAUDE.md
Claude Code instructions for memi-design/memi, covering mémoire — guidelines for claude, prime directives, what is mémoire, widget v2 operating model and architecture.
memi AGENTS.md
AGENTS.md instructions for memi-design/memi, covering mémoire — guidelines for codex, prime directives, what is mémoire, architecture and atomic design levels.
GameStudio-CodexKIT AGENTS.md
AGENTS.md instructions for hoatv2211/GameStudio-CodexKIT, covering gamestudio-codexkit operating contract, active sources of truth, primary distribution, repository-local maintenance and evidence.
chiplab copilot-instructions.md
Instructions for veecle/chiplab: Agent instructions for this repository live in AGENTS.md at the repo root. Please read that file for repository conventions, available workflows, and guidelines.
codev CLAUDE.md
Instructions for cluesmith/codev, covering codev project instructions for ai agents, always-on engineering context (hot tier), this repository is codev, built with codev, how framework files resolve and gates.