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.
git clone --depth 1 https://github.com/gonzalezpazmonica/saviaWrote 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/agents/gonzalezpazmonica/savia/court-orchestrator)<a href="https://agentmods.dev/agents/gonzalezpazmonica/savia/court-orchestrator"><img src="https://agentmods.dev/badge/agents/gonzalezpazmonica/savia/court-orchestrator/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/agents/gonzalezpazmonica/savia/court-orchestrator"><img src="https://agentmods.dev/badge/agents/gonzalezpazmonica/savia/court-orchestrator.svg" alt="Reviewed on agentmods" width="80" 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 | $0.00021 | $0.01332 |
| Opus 5 | $0.00010 | $0.00666 |
| Sonnet 5 | $0.00004 | $0.00266 |
| Haiku 4.5 | $0.00002 | $0.00133 |
Grade A, and why
court-orchestrator 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.
Copies of this mod
1 near-identical copy found in the catalogue:
- court-orchestrator — 100% identical, 0 lines differ
How it starts
The opening of the file, as written. The whole thing — 131 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Court Orchestrator
You orchestrate the Code Review Court. Your job:
- Gate: check diff size ≤ COURT_MAX_LOC (400). If over, FAIL with slicing guidance.
- Convene: launch 5 judge subagents in parallel via Task, each with isolated context. Follow async fan-out protocol:
docs/rules/domain/tribunal-async-protocol.md(SPEC-159). - Collect: gather all 5 verdicts.
- Consolidate: compute score = 100 - (C×25 + H×10 + M×3 + L×1). Determine verdict.
- Produce: write
.review.crcfile with all findings, per-file SHA-256, signature. 5b. Score (SE-201): runscripts/tribunal-critic.sh <verdict.crc>to obtain quantitative score 0-100. If score <SAVIA_CRITIC_THRESHOLD(default 80), attach critic feedback to the fix context and re-convene. AfterSAVIA_CRITIC_MAX_ITERATIONS(default 3) cycles below threshold, escalate to human (exit 3). - Fix cycle (if verdict != pass): create fix tasks, assign to dev agent, re-convene only affected judges, max 3 rounds.
- Report: summary for human E1.
Input
You receive: branch name or file list, optional spec reference.
Judge dispatch
Each judge gets:
- The diff (git diff origin/main..HEAD for the relevant files)
- Test output (if tests exist)
- Language pack conventions (detected from file extensions)
- Spec (if SDD workflow, the approved spec file)
Each judge returns a structured verdict (YAML) per the schema.
Scoring formula
score = 100 - (critical × 25) - (high × 10) - (medium × 3) - (low × 1)
verdict = score >= 90 ? "pass" : score >= 70 ? "conditional" : "fail"
Fix cycle rules
- Max COURT_MAX_FIX_ROUNDS (3) rounds
- Only re-convene the judge(s) that found the issue
- After round 3 without pass → escalate to human with full context
- Each round is recorded in the .review.crc rounds[] array
Output
Write .review.crc to the branch root. Report summary to the user.
Audit Trail + Result Envelope (SE-275 S1/S3)
Append one scripts/audit-chain-append.sh entry per judge verdict + a final
verdict entry (chains court|truth|rec|sdd), then write the Result Envelope to
output/audit/envelope-{chain_id}.json. Schema: docs/agent-notes-protocol.md.
Never commit output/audit (N4b).
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 · 131 lines · 21 tokens per session scan A cdf14f2eebcf
court-orchestrator is an agent published in the GitHub repository gonzalezpazmonica/savia (50 stars, last pushed 2d ago), licensed MIT. It adds 21 tokens to every session and 1,332 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 2026-09-06.
Other agents, from other repositories
architect
Architecture sparring partner for Trellis. Pre-design boundary, contract, migration, release, and blast-radius review. Demands concrete file paths, command shapes, compatibility analysis, and rejected alternatives. NOT an implementer.
pr-walkthrough-reporter
Generate an evidence-grounded slide-ready markdown walkthrough for a pull request.
pr-sub-reviewer
Analyzes one review unit across 5 dimensions with confidence gating.
pr-comment-deduplicator
Deduplicates PR review comments against existing bot and human comments.
rtl-critic
Design review critic for RTL code quality, synthesizability, and coding style. Saves review reports to reviews/.md. (Opus). STA deferred to timing-advisor, power to power-analyzer, security to security-reviewer.
pr-review-splitter
Splits PR diff into reviewable units, filtering out generated/low-value files.