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 commands/terrene-foundation/metis/codifygit clone --depth 1 https://github.com/terrene-foundation/metisWhat 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 | $0.00025 | $0.01572 |
| Opus 5 | $0.00013 | $0.00786 |
| Sonnet 5 | $0.00005 | $0.00314 |
| Haiku 4.5 | $0.00003 | $0.00157 |
Grade A, and why
codify 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.
How it starts
The opening of the file, as written. The whole thing — 136 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Workspace Resolution
- If
$ARGUMENTSspecifies a project name, useworkspaces/$ARGUMENTS/ - Otherwise, use the most recently modified directory under
workspaces/(excludinginstructions/) - If no workspace exists, ask the user to create one first
- Read all files in
workspaces/<project>/briefs/for user context (this is the user's input surface)
Phase Check
- Read
workspaces/<project>/04-validate/to confirm validation passed - Read
docs/anddocs/00-authority/for knowledge base - Output: update existing agents and skills in their canonical locations (e.g.,
agents/frameworks/,skills/01-core-sdk/,skills/02-dataflow/, etc.)
Execution Model
This phase executes under the autonomous execution model (see rules/autonomous-execution.md). Knowledge extraction and codification are autonomous — agents extract, structure, and validate knowledge without human intervention. The human reviews the codified output at the end (structural gate on what becomes institutional knowledge), but the extraction and synthesis process is fully autonomous.
Workflow
1. Consume learning digest
Before extracting new knowledge, integrate what the learning system has captured:
- Read
.claude/learning/learning-digest.json— the structured summary of recent observations - Read
.claude/learning/learning-codified.json— what was previously codified (avoid re-processing) - Read recent journal entries referenced in the digest (
decisionsarray) — DECISION and DISCOVERY entries contain semantic context - Read
.session-notes— latest session accomplishments and outstanding items
Analyze the digest for actionable findings:
- Corrections → Do any rules or skills need updating to match user preferences? Each correction is a real signal where the user pushed back on an approach.
- Error patterns → Should any recurring rule violations become new rule sections (DO/DO NOT with examples)?
- Decisions → Should any architectural decisions from journals become agent or skill knowledge?
- Accomplishments → Do any completed features need documentation in skills?
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 · 136 lines · 25 tokens per session scan A cc19fce69e67
codify is a command published in the GitHub repository terrene-foundation/metis (2 stars, last pushed 4mo ago), licensed Apache-2.0. It adds 25 tokens to every session and 1,572 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.
Other commands, from other repositories
learn
Force claude-smart to extract learnings from this session now.
component
Scaffold a new React component grounded in the paper-mono primitives. Requires explicit kind or a nearest-existing-component match. No empty divs, no speculative scaffolding.
memory-store
Store an insight, decision, or pattern to memory.
review
Cold re-quiz on code that already shipped — your own session commits, not the change in front of you.
no-vibe
Enter no-vibe mode in OpenCode (tutor mode, no direct project file writes).
teach-me-testing
Teach testing progressively through structured sessions. Use when user says ""lets learn testing"" or ""I want to study test practices"".