concilium-forge

concilium-forge is a command for Claude Code from raichominev/concilium. It costs 29 tokens per session (676 once invoked), scanned A, original, Apache-2.0.

A command for running a collaborative idea-generation round with several AI models. The models read a shared register of ideas and build on one another without judging or voting.

In plain words
What is it for?
Use it to explore an open question, record what has already been tried, develop original approaches, and organize the resulting ideas.
Why use it?
It separates generating possibilities from evaluating them, so early ideas can develop before being rejected.

Command for Claude Code

Written for Claude Code: argument-hint in frontmatter.

Part of the concilium plugin — 1 skill, 3 commands shipped together

Good fit Use it to explore an open question, record what has already been tried, develop original approaches, and organize the resulting ideas.

Compare 6 commands from other repositories ↓
Install with agentmods
npx agentmods add commands/raichominev/concilium/concilium-forge
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.

Clone the repo
git clone --depth 1 https://github.com/raichominev/concilium

Made for: Claude Code.

Or install concilium, the plugin that ships this one along with the rest of its 1 skill, 3 commands.

Wrote 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.

agentmods badge for concilium-forge

README.md
[![agentmods](https://agentmods.dev/badge/commands/raichominev/concilium/concilium-forge.svg)](https://agentmods.dev/commands/raichominev/concilium/concilium-forge)
Your own site
<a href="https://agentmods.dev/commands/raichominev/concilium/concilium-forge"><img src="https://agentmods.dev/badge/commands/raichominev/concilium/concilium-forge.svg" alt="Measured on agentmods" height="20"></a>
Per session 29 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 676 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. A grade says what 26 rules found in the file — not that it is safe.
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.

ModelPer sessionOnce invoked
Fable 5.1 $0.00029 $0.00676
Opus 5 $0.00015 $0.00338
Sonnet 5 $0.00006 $0.00135
Haiku 4.5 $0.00003 $0.00068

Measured 7d ago against content hash 55696abe60b4, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-07, from the pricing page.

Security

Grade A, and why

concilium-forge 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 7d 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.

commands/concilium-forge.md · 48 lines

How it starts

The opening of the file, as written. The whole thing — 48 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Run a forge round on: $ARGUMENTS

Forge is the generative mode and the opposite discipline to /concilium-review. Seats produce ORIGINAL ideas, read each other's from a shared register, and BUILD on them. No verdicts, no five blocks, no ratification, and no seat judges another seat — the duty to refute kills a half-formed idea before anyone can extend it. Read references/forge-mode.md before the first round.

If $ARGUMENTS is a question (new forge)

  1. Write a brief from references/forge-brief-template.md. Three blocks carry it: the QUESTION, the nearby inventory (what the project already has to hand), and ALREADY TRIED — the last is worth more than any instruction, because it stops seats scoring on dead arms and pushes them into the empty cells.
  2. Round 1, no register, seats in parallel — background, ~10 min timeout each: scripts/concilium-forge.sh <seat> 1 --brief BRIEF.md --out <dir> Give every seat a fresh empty working directory. Measured: seats sharing a directory read each other's output and position against it instead of thinking, which voids any claim that they converged independently.
  3. Curate into a register: OPEN IDEAS (clustered, ranked by consequence) → HISTORY (with reuse counts) → EXPERIMENT LOG. Merge duplicates, credit every contributing id, and add your own ideas — the orchestrator is a seat too, including for the uncomfortable meta-ones.
  4. Round 2 with --register <file>. Seats must name the ids they built on. This is where the value appears: round 1 tends to converge, round 2 produces the reframings.
  5. Re-curate, then name the gate — the one experiment whose result orders all the others — and run it. A forge that never executes an experiment is a brainstorm.

If $ARGUMENTS starts with --continue

Read the existing register, run the next round with --register, re-curate. Move anything acted upon into HISTORY; keep OPEN IDEAS on top.

Two rules that cost a round each to learn

Read the full file on GitHub · 48 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.

  1. 7d ago First seen · 48 lines · 29 tokens per session scan A 55696abe60b4

Subscribe to this mod's changes

concilium-forge is a command published in the GitHub repository raichominev/concilium (2 stars, last pushed 13d ago), licensed Apache-2.0. It adds 29 tokens to every session and 676 once invoked, about $0.0001 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 2026-08-31.