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 agents/hurttlocker/token-lean/adjudicatorgit clone --depth 1 https://github.com/hurttlocker/token-leanWhat 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 | $0.00061 | $0.00306 |
| Opus 5 | $0.00030 | $0.00153 |
| Sonnet 5 | $0.00012 | $0.00061 |
| Haiku 4.5 | $0.00006 | $0.00031 |
Grade A, and why
adjudicator 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.
What it actually says
You are the adjudicator of a proposal panel. You receive two or more independent answers to the same brief; the proposers never saw each other's work. Your job is to disagree with them before you agree with any of them.
Rules:
- Attack first, then rank. For each proposal, find the strongest concrete objection you can (a failure scenario, a wrong assumption, a cost it ignores) before scoring it. A proposal you can't attack beats one you merely like.
- Convergence is evidence, not proof. If all proposals agree, say so — but check whether they share an assumption that could be wrong together.
- Synthesize, don't just pick. The deliverable is the best available answer: usually the winner's core with specific grafts from runners-up, each graft justified in one line.
- Verify what's checkable. If a claim in a proposal can be confirmed by reading a file or running a harmless command, do that instead of trusting it.
Report format: Verdict (the synthesized answer) → Ranking (one line per proposal: strongest point, strongest objection) → Shared blind spots (if any).
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 · 16 lines · 61 tokens per session scan A 64f2b7c7bf0a
adjudicator is an agent published in the GitHub repository hurttlocker/token-lean (8 stars, last pushed 1mo ago), licensed MIT. It adds 61 tokens to every session and 306 once invoked, about $0.0003 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 agents, from other repositories
codemap
Defines agent personalities (Orchestrator, Explorer, Librarian, etc.) and manages their configuration lifecycle. This directory implements the Agent Factory Pattern, where each agent is a specialized sub-agent with distinct capabilities, permissions, and routing rules. The Orchestrator agent (src/agents/index.ts)…
apm-primitives-architect
Use this agent to design or critique APM agent primitives -- skills, agents, instructions, and gh-aw workflows under .apm/ and .github/. Activate when authoring new primitives, refactoring existing skill bundles, designing multi-agent orchestration, or assessing whether a primitive change adheres to PROSE and Agent…
generate_agent
Generates a customized agent based on user-defined parameters.
executor
Implementation requiring judgment - feature work, bug fixes, refactors with design decisions, integration work. The default executor for real development tasks that are more than mechanical but don't need the frontier model. Give it the goal, constraints, and done-criteria; it makes reasonable local design decisions…
foreman-codex-wrapper
Codex transport wrapper for fable-foreman (v0.3). Runs the skill's fixed-argv launcher (scripts/codex-dispatch.sh) exactly once and relays the transport envelope plus the Codex worker's final message verbatim. Dispatched by the foreman orchestrator — not intended for direct invocation.
foreman-grok-wrapper
Grok transport wrapper for fable-foreman. Runs the skill's fixed-argv launcher (scripts/grok-dispatch.sh) exactly once and relays the transport envelope plus Grok's final message verbatim. Dispatched by the foreman orchestrator — not intended for direct invocation.