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/armanfatemi/nulliusWrote 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/armanfatemi/nullius/comply)<a href="https://agentmods.dev/commands/armanfatemi/nullius/comply"><img src="https://agentmods.dev/badge/commands/armanfatemi/nullius/comply/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/commands/armanfatemi/nullius/comply"><img src="https://agentmods.dev/badge/commands/armanfatemi/nullius/comply.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 | $0.00033 | $0.01820 |
| Opus 5.5 | $0.00013 | $0.00728 |
| Sonnet 5.5 | $0.00007 | $0.00364 |
| Haiku 4.5 | $0.00003 | $0.00182 |
Grade A, and why
comply 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 1mo 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 — 154 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Rule compliance is checked in deterministic steps around one dispatch step. The protocols live in the CLI so they cannot drift from the checker. Emit them and follow them exactly.
Determine the touch-list
$ARGUMENTS names either a plan and lets its touch-list be derived, or the
touch-list itself:
- If
$ARGUMENTSis one or more repo-relative paths (no spaces, each ending in a known extension), use them directly as the touch-list. - If
$ARGUMENTSnames one markdown file, treat it as the plan and derive its touch-list from every backticked, repo-relative path with a known extension (.ts,.md,.json,.jsonl,.yml/.yaml,.sh) that it mentions — the same conventionnullius-kit pipeline touched-areasuses to derive a proposal's own touched paths. - If
$ARGUMENTSis empty, use the most recently modified.mdfile in the plans directory (.claude/plansor~/.claude/plans, newest first) as the plan, and derive its touch-list the same way.
Call the touch-list <paths...> below and the plan (if one was used) <plan>.
1. Select the rules that bind
npx -y @nullius-inverba/claims rules select --paths <paths...>
Deterministic, no model involved: it prints the id of every rule under
.claude/rules/ whose applies_to matches at least one path in the
touch-list, then the excluded count. Keep the excluded count — report it to
the user later even if it is zero, because a selection that silently narrows
is the failure this verb exists to prevent.
If zero rules are selected, say so and stop: there is nothing to dispatch, and that is a real answer, not a failure to find work.
2. Dispatch one starved brief per selected rule
For each rule id rules select printed:
npx -y @nullius-inverba/claims rules select --paths <paths...> --emit-brief <rule-id>
This prints one starved compliance brief on stdout, and nothing else — the rule's id and full text, plus the touch-list, with no sibling rules, no plan rationale, no narrative.
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.
- 1mo ago First seen · 154 lines · 33 tokens per session scan A 7a32bafe2cfd
comply is a command published in the GitHub repository armanfatemi/nullius (5 stars, last pushed today), licensed MIT. It adds 33 tokens to every session and 1,820 once invoked, about $0.0001 per session on Opus 5.5. 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
check
Fail on NET-NEW architecture violations only — the same ratchet ArchSteer runs in CI, run right now.
review-branch
Review the current branch's diff against base by dispatching atomic-reviewer. No orchestration loop, no spec required — pre-flight before /commit pr or /commit merge.
advanced-code-review-context
Advanced Code Review Phase 2: Context Analysis - load previous reviews, PR history, declined items.
code-review-give
Give mode for code-review: Review someone else's code with multi-pass analysis and structured recommendations.
mach12:issue-review
Read a GitHub issue and all comments, review the implementation plan, and present findings.
review-design-verify
Phases 4-5 of reviewing-design-docs: Interface Verification + Implementation Simulation.