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/aaronbassett/agent-foundry/tribunalgit clone --depth 1 https://github.com/aaronbassett/agent-foundryWrote 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.
[](https://agentmods.dev/commands/aaronbassett/agent-foundry/tribunal)<a href="https://agentmods.dev/commands/aaronbassett/agent-foundry/tribunal"><img src="https://agentmods.dev/badge/commands/aaronbassett/agent-foundry/tribunal.svg" alt="Measured on agentmods" height="20"></a>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 5.1 | $0.00040 | $0.01306 |
| Opus 5 | $0.00020 | $0.00653 |
| Sonnet 5 | $0.00008 | $0.00261 |
| Haiku 4.5 | $0.00004 | $0.00131 |
Grade A, and why
tribunal 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 5d 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 — 104 lines — stays where its author put it; the contents beside it link to each section on GitHub.
/decision-making:tribunal
When to use
Use tribunal when you have N≥2 concrete options on the table, the decision is load-bearing and hard to reverse, and you want rigorous deliberation — including rebuttals, judge questions, and a pre-judgment disclosure — before committing. Tribunal is heavier than other techniques in the plugin because it spawns parallel advocates, runs multiple rounds, and interrogates its own conclusions. Pay that cost only when the decision warrants it.
Escalate from adversarial to tribunal when the stakes justify the extra cost: irreversible migrations, architecture choices that lock in future work, hiring and team-shape decisions, vendor lock-in, or anything you will have to live with for a long time. Tribunal is the only technique in v1 that offers an honest "unable to decide" escape hatch — if the options are genuinely equivalent, or if the decision depends on information you do not have, the tribunal can and should say so rather than fabricating a verdict.
Cost tier
High. Up to 10 rounds × N advocates × opus model + fact-finding phase. Gate behind a condition when referencing from plan steps. See references/cost-tiers.md.
Input
A decision description and optional list of options. If options are unclear, enter fact-finding and ask 3-5 clarifying questions before spawning advocates.
Workflow
-
Parse the decision — identify what is being decided, what the options are (minimum 2, no maximum), what criteria matter, and what context is relevant. If options are unclear, ask the user to enumerate them explicitly and confirm you have understood. Output: a clear list of N options to deliberate.
-
Fact-finding — probe for constraints (budget, timeline, technical limitations), context (why now, what prompted it), history (what has been tried, what has been ruled out), stakeholders (who else is affected, what are their preferences), non-negotiables (must-haves vs nice-to-haves), and success criteria (how will you know the decision was correct). Ask 3-5 clarifying questions. Conclude fact-finding when you understand the decision's importance and urgency, know the key constraints advocates must work within, and could explain the decision context to a colleague. More than 10 questions suggests the decision isn't ready for deliberation.
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.
- 5d ago First seen · 104 lines · 40 tokens per session scan A dfc65020f5cb
tribunal is a command published in the GitHub repository aaronbassett/agent-foundry (4 stars, last pushed 20d ago), licensed MIT. It adds 40 tokens to every session and 1,306 once invoked, about $0.0002 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
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.
converge
Assess the current codebase against the feature's spec, plan, and tasks, then append any remaining unbuilt work as new tasks to tasks.md so implement can complete it.
implement
Execute the implementation plan by processing and executing all tasks defined in tasks.md.
analyze
Perform a non-destructive cross-artifact consistency and quality analysis across spec.md, plan.md, and tasks.md after task generation.