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/implementgit 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 | $0.00020 | $0.01848 |
| Opus 5 | $0.00010 | $0.00924 |
| Sonnet 5 | $0.00004 | $0.00370 |
| Haiku 4.5 | $0.00002 | $0.00185 |
Grade A, and why
implement 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 — 134 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Workspace Resolution
- If
$ARGUMENTSspecifies a project name or todo, parse accordingly - 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 files in
workspaces/<project>/todos/active/to see what needs doing - Read files in
workspaces/<project>/todos/completed/to see what's done - If
$ARGUMENTSspecifies a specific todo, focus on that one - Otherwise, pick the next active todo
- Reference plans in
workspaces/<project>/02-plans/for context - If any file in
briefs/was modified afterspecs/_index.md, STOP — briefs changed since analysis. Flag for user decision: re-run/analyzeor acknowledge the brief change.
Execution Model
This phase executes under the autonomous execution model (see rules/autonomous-execution.md). Implementation is fully autonomous — agents execute in parallel, self-validate through TDD, and converge through quality gates. The human observes outcomes but does not sit in the execution loop. Pre-existing failures are fixed, not reported (zero-tolerance). Agent-to-agent delegation (reviewer, security-reviewer) is autonomous, not human-gated.
Workflow
NOTE: Run /implement repeatedly until all todos/active have been moved to todos/completed
1. Prepare todos
You MUST always use the todo-manager to create detailed todos for EVERY SINGLE TODO in todos/000-master.md.
- Review with agents before implementation
- Ensure that both FE and BE detailed todos exist, if applicable
2. Context anchoring (MUST run before each todo)
Before implementing ANY todo, re-read the source material that spawned it:
- Re-read the relevant spec files — check
specs/_index.md, identify which spec files cover the domain of this todo, and read them. The spec is the authority on what to build. If the spec sayscreate_user(name, email, password_hash), that is the signature. - Re-read the plan section in
02-plans/that this todo implements — not the whole directory, but the specific plan paragraphs. If the plan describes aDataFabricclass with 3 methods, you are building that class with those 3 methods. - Re-read relevant journals in
workspaces/<project>/journal/— decisions, trade-offs, and risks from analysis inform how to implement. If a journal says "chose event-driven over polling because of X," the implementation must be event-driven. - Re-read the todo itself — the description, not just the title. Todos have implementation details that get ignored when agents skim titles.
- Read current source code before calling any existing service or function. Do not trust plans, specs, or previous todos for current method signatures — the code is the truth for what exists NOW. Specs are the truth for what SHOULD exist.
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 · 134 lines · 20 tokens per session scan A d10cddbe092d
implement is a command published in the GitHub repository terrene-foundation/metis (2 stars, last pushed 4mo ago), licensed Apache-2.0. It adds 20 tokens to every session and 1,848 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
git
Git operations with intelligent commit messages and workflow optimization.
checklist
Generate a custom checklist for the current feature based on user requirements.
clarify
Identify underspecified areas in the current feature spec by asking up to 5 highly targeted clarification questions and encoding answers back into the spec.
specify
Create or update the feature specification from a natural language feature description.
analyze
Perform a non-destructive cross-artifact consistency and quality analysis across spec.md, plan.md, and tasks.md after task generation.
constitution
Create or update the project constitution from interactive or provided principle inputs.