What is contentforge AGENTS.md?

Project instructions for ContentForge, an open-source system that turns topics into fact-checked, brand-compliant Microsoft Word documents and manages their publication lifecycle. They explain the project’s agent context, entry points, and important files.

Use them when developing or operating ContentForge, including its document-generation, fact-checking, publishing, measurement, auditing, and content-planning workflows.

They give coding agents the background needed to work consistently in the ContentForge repository. Without that context, an agent may miss the project’s content, quality, or compliance processes.

contentforge AGENTS.md — Instructions file by indranilbanerjee

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/indranilbanerjee/contentforge/agents-md

Clone the repo

git clone --depth 1 https://github.com/indranilbanerjee/contentforge

Made for: Codex, OpenCode.

Per session 1,817 This file is loaded in full into every session.

When invoked 1,817 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	$0.01817	$0.01817
Opus 5	$0.00908	$0.00908
Sonnet 5	$0.00363	$0.00363
Haiku 4.5	$0.00182	$0.00182

Measured 3d ago against content hash 20356fe80ff2, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

contentforge AGENTS.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 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.

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.

Origin

Copies of this mod

1 near-identical copy found in the catalogue:

ContentForge-techshu AGENTS.md — 97% identical, 2 lines differ

AGENTS.md · 64 lines

How it starts

The opening of the file, as written. The whole thing — 64 lines — stays where its author put it; the contents beside it link to each section on GitHub.

ContentForge — agent context

This file is auto-loaded by OpenAI Codex, Google Antigravity, GitHub Copilot CLI, Cursor (when in the agent context chain), and other Agent Skills runtimes. Equivalent of CLAUDE.md for non-Claude surfaces.

What this plugin is

ContentForge is an open-source content lifecycle system — turn a one-line topic into a publication-ready, fact-checked, brand-compliant Microsoft Word document (.docx with C2PA content provenance signing for EU AI Act Article 50 compliance) in 30–60 minutes, then keep the library alive: published pieces are measured (cf-aeo-check history), audited for decay (cf-audit, recorded durably via scripts/audit-ledger.py), and fed back into the next calendar and brief through file contracts that survive sessions. 22 skills · 13 specialist agents · 8 built-in content types · 10 quality gates · 43-pattern AI-detection humanizer · fact-checker subagent · three-category internal linking · author/E-E-A-T byline layer · machine-readable pipeline contract (config/pipeline-graph.json) · cross-run telemetry with advisory floors (scripts/telemetry.py).

Supported surfaces (v4.1.2): Claude Code (CLI + IDE extensions), Anthropic Cowork, OpenAI Codex (CLI + IDE + App), Cursor 2.5+, GitHub Copilot CLI, Google Antigravity 2.0 (CLI + IDE), Hermes Agent, OpenClaw, Grok (xAI Build CLI, via the native .grok-plugin/ pair). Plus 35+ additional platforms via the Agent Skills open standard, any Agent Plugins 1.0 host via the root plugin.json (with the portable execution lane for hosts without subagent dispatch), and five hero skills as claude.ai-uploadable .skill release assets.

How to use it as an agent

Discover skills by description. All 22 skills auto-discover via SKILL.md frontmatter (name: + description:). Match user intent.
Pipeline order matters. ContentForge assumes the canonical 10-phase pipeline plus Step 0.5: Step 0.5 title curation → Step 0 Client Site Reconnaissance (brand facts + Internal-Link Inventory, required when the brand has a website) → Phase 1 research → Phase 2 fact-check → Phase 3 drafting → Phase 4 scientific validation → Phase 5 structuring/proofreading → Phase 6 SEO/AEO/GEO optimization (deep-link rule + thin-brand_pages guard) → Phase 6.5 humanizer (43-pattern AI-detection sweep, grounding-first) → Phase 7 reviewer scorecard → Phase 8 output (.docx export). Social adaptation and translation run post-pipeline. Reviewer is Phase 7; Output Manager is Phase 8.
Skill bodies reference Python scripts at scripts/<name>.py — invoke via Bash / run_shell_command. Critical scripts: generate-docx.py (with C2PA signing flags), checkpoint-manager.py, plugin-metadata.py, connector-status.py, harvest-brand-pages.py (stdlib, robots-respecting site crawler used by brand-setup), text-metrics.py (burstiness/FK/placement gates plus --ai-tell-scan). Humanization and fact-checking are agent behaviors defined in agents/ and config/humanization-patterns.json — there are no humanizer.py or fact-checker.py scripts.
HTTP MCP connectors are opt-in. Full catalog at .mcp.json.connectors-reference (Slack, Notion, Canva, Webflow, Gmail, Google Calendar, Figma + 9 more). Users opt in by configuring env vars or MCP servers.
C2PA content provenance signing for the .docx is required for EU distribution. Article 50 enforcement starts 2 Aug 2026. --c2pa-sign flag on scripts/generate-docx.py.
AI-detectability is advisory, never a gate. text-metrics.py --ai-tell-scan is a deterministic, dependency-free proxy scan (aphorism density, banned lexemes, connective/participial openers, uniform runs) that reports a LOW/MODERATE/HIGH rating, surfaced in the Phase 6.5 report, the reviewer's Readability sub-score, and the Completion Card. It never blocks publication and the pipeline never claims to "beat" any specific detector — see references/ai-detection-signals.md for the reasoning it's built on.

Read the full file on GitHub · 64 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.

3d ago First seen · 64 lines · 1,817 tokens per session scan A 20356fe80ff2

Subscribe to this mod's changes

contentforge AGENTS.md is an instructions file published in the GitHub repository indranilbanerjee/contentforge (26 stars, last pushed 16d ago), licensed MIT. It adds 1,817 tokens to every session, about $0.0091 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.

Documents Writing and docs aeo claude-code claude-plugin content-pipeline fact-checking seo

digital-marketing-pro AGENTS.md

Instructions for indranilbanerjee/digital-marketing-pro, covering digital marketing pro — agent context, what this plugin is, how to use it as an agent, canonical entry points and files in this repo.

indranilbanerjee/digital-marketing-pro · 2,038 tokens

agent-skills GEMINI.md

Instructions for conversiontools/agent-skills, covering conversion tools gemini extension, authentication, usage and more.

conversiontools/agent-skills · 519 tokens

vscode buildNext.instructions.md

Working notes and architecture documentation for the new esbuild-based build system in build/next. Use when making changes to the new build pipeline (transpile/bundle commands, NLS plugin, source-map handling, resource copying, or self-hosting watch tasks).

microsoft/vscode · 6,785 tokens

contentforge AGENTS.md