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 skills add mhashim6/kauri --skill proposegit clone --depth 1 https://github.com/mhashim6/kauriWrote 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/skills/mhashim6/kauri/propose)<a href="https://agentmods.dev/skills/mhashim6/kauri/propose"><img src="https://agentmods.dev/badge/skills/mhashim6/kauri/propose/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/skills/mhashim6/kauri/propose"><img src="https://agentmods.dev/badge/skills/mhashim6/kauri/propose.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 | $0.00119 | $0.00741 |
| Opus 5.5 | $0.00048 | $0.00296 |
| Sonnet 5.5 | $0.00024 | $0.00148 |
| Haiku 4.5 | $0.00012 | $0.00074 |
Grade A, and why
propose 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 — 72 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Propose a Decision
Draft a decision record and present it for human approval before writing it to the database. This is the quality gate that prevents low-value or poorly-worded records from accumulating.
Steps
-
Understand the context: Based on
$ARGUMENTSand the conversation so far, identify what decision was made, why, and what alternatives were considered. -
Check for duplicates: Run
kauri_query --text "$ARGUMENTS"to see if a similar decision already exists. If it does, ask whether to supersede it or skip recording. -
Check the taxonomy: Run
kauri_taxonomy_listto see available tags. Pick the most fitting ones. -
Draft the record: Present the following to the user for review:
Title: [short, scannable — what was decided] Body: [What was decided and why] [What alternatives were considered] [Key constraints or trade-offs] Tags: [tag1, tag2] Files: [any associated file paths] Links: [IDs of related decisions, if any] Pin: [yes/no — only if this is a critical rule the agent must always see] Status: [active or draft if still tentative] -
Ask for approval: Present the draft and ask:
- "Record this decision?" (proceed)
- "Edit something?" (revise and re-present)
- "Skip" (don't record)
-
Record if approved: Call
kauri_recordwith the approved values. If superseding, include--supersedes <id>.
Quality guidelines
A good decision record:
- Title: One sentence, starts with a verb or states the choice clearly. Good: "Use JWT with 15-minute refresh tokens". Bad: "Auth stuff".
- Body: Explains WHY, not just WHAT. Includes the trade-off or constraint that drove the choice.
- Tags: 1-3 tags from the taxonomy. Don't force-fit — if nothing matches, propose a new tag and explain why.
- Files: Only files directly governed by this decision. Don't list every file that touches the feature.
- Pin: Reserve for critical rules (security constraints, hard boundaries). Most decisions should NOT be pinned.
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 · 72 lines · 119 tokens per session scan A ef86685ad81c
propose is a skill published in the GitHub repository mhashim6/kauri (5 stars, last pushed 5mo ago), licensed MIT. It adds 119 tokens to every session and 741 once invoked, about $0.0005 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 skills, from other repositories
agent-state-bar
A design guide for an agent status bar: a short, automatically updated summary of progress, remaining tasks, tool counts, errors, and system state placed near the latest model input.
context-engineering
A guide to organizing the instructions and information an AI agent receives at each step. It covers system prompts, tool descriptions, conversation history, on-demand skills, and protections against unsafe outside content.
kv-cache-design
A guide to arranging an AI agent's conversation so previously processed text can be reused. A KV cache stores intermediate results for earlier tokens, while a prompt cache reuses a stable beginning across requests.
reward-design
A guide to defining the scores that an AI agent receives during trial-and-error training. It covers checks based on rules, human preferences, or another model, and distinguishes rewarding the final result from rewarding the steps taken.
agent-evaluation
A guide to measuring whether an AI agent works reliably on the tasks that matter. It covers repeatable test environments, program checks, human-like scoring rules, failure analysis, and experiments that separate model problems from system problems.
ai-style
A Chinese-language writing guide for producing or revising reader-facing text such as product announcements, public-account articles, emails, and README files. It focuses on reducing patterns that make writing sound machine-generated.