What is forge CLAUDE.md?

Repository instructions for working on Forge, a multi-module Go project with core libraries, a command-line tool, and channel plugins.

Use it when changing Forge code, formatting modules, running pre-commit checks, or implementing platform and MCP integrations.

They give coding agents the project layout, required code checks, and rules for connecting Forge to external platforms. This reduces mistakes such as using the wrong module or omitting required authentication headers.

forge CLAUDE.md — Instructions file by initializ

Install

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.

agentmods

npx agentmods add instructions/initializ/forge/claude-md

Clone the repo

git clone --depth 1 https://github.com/initializ/forge

Per session 605 This file is loaded in full into every session.

When invoked 605 The same file — it is already loaded in full.

Security scan A 0 findings. Scan, not verified.

Origin original No closer match found in the catalogue.

Token cost

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	$0.00605	$0.00605
Opus 5	$0.00302	$0.00302
Sonnet 5	$0.00121	$0.00121
Haiku 4.5	$0.00060	$0.00060

Measured 2d ago against content hash 7ae7cf1c9ae7, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

forge CLAUDE.md 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 2d 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.

CLAUDE.md · 42 lines

How it starts

The opening of the file, as written. The whole thing — 42 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Forge - CLAUDE.md

Project Structure

Multi-module Go workspace with three modules:

forge-core/ — Core library (registry, tools, security, channels, LLM)
forge-cli/ — CLI commands, TUI wizard, runtime
forge-plugins/ — Channel plugins (telegram, slack), markdown converter

Platform-integration contracts (used by initializ AIP)

Tenancy headers: every forge→platform HTTP callout (admission, remote session store, MCP platform token resolver) MUST send Org-Id + Workspace-Id from FORGE_ORG_ID/FORGE_WORKSPACE_ID env alongside Authorization: Bearer ${FORGE_PLATFORM_TOKEN} — the platform verifies a PER-ORG HS256 token and needs Org-Id to pick the signing secret BEFORE it can validate the bearer (omitting it → 401 "missing org-id header"). Pattern lives in admission_loader.go / remote_session_store.go / mcp/platform_token.go.
MCP auth types (auth.type): oauth/bearer/static + managed platform (agent-principal token from ForgeConfig.Platform.token_endpoint) and user (delegated, lazy — cannot be required:true). The platform owns the op→identity split; forge only reads per-server config.
DEFER approval is FULLY BUILT & enforced (forge-core/security/deferpolicy/, Slack Block-Kit delivery in forge-plugins/channels/slack/approvals.go): security.defer.tools.<toolName>{to: channel:slack:#x, approvers, timeout} — keyed by the RUNTIME tool name, which for MCP is <server>__<op>. Guardrails approvalGates are parsed but NOT enforced — use security.defer.*.
MCP OAuth discovery/DCR (forge-core/mcp/oauth_discovery.go, #316/#320): RFC 9728→8414→7591. wellKnown() must INSERT the well-known segment for path-qualified issuers (Atlassian auth.atlassian.com/<id>) — replacing the path resolves the wrong AS.

Pre-Commit Requirements

Always run before committing:

# Format all modules
gofmt -w forge-core/ forge-cli/ forge-plugins/

# Lint all modules
golangci-lint run ./forge-core/...
golangci-lint run ./forge-cli/...
golangci-lint run ./forge-plugins/...

Read the full file on GitHub · 42 lines

Changes

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.

2d ago First seen · 42 lines · 605 tokens per session scan A 7ae7cf1c9ae7

Subscribe to this mod's changes

forge CLAUDE.md is an instructions file published in the GitHub repository initializ/forge (156 stars, last pushed 5d ago), licensed Apache-2.0. It adds 605 tokens to every session, about $0.0030 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 2026-08-30.

Git and workflow Languages Planning a2a a2a-protocol agent-cli ai-agents ai-agents-cli ai-workforce

forge CLAUDE.md

Getting it into your agent

What it costs to keep this loaded

Grade A, and why

Nothing flagged

How it starts

Forge - CLAUDE.md

Project Structure

Platform-integration contracts (used by initializ AIP)

Pre-Commit Requirements

What this file has done since we first saw it

Other instructions, from other repositories

trpc-agent-go AGENTS.md

a2a-python AGENTS.md

lucid-agents AGENTS.md

agentarea AGENTS.md

adk CLAUDE.md

a2a-editor CLAUDE.md