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/morpho-org/sdks/skill-authoringgit clone --depth 1 https://github.com/morpho-org/sdksWhat 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 | $0.00003 | $0.01554 |
| Opus 5 | $0.00002 | $0.00777 |
| Sonnet 5 | $0.00001 | $0.00311 |
| Haiku 4.5 | $0.00000 | $0.00155 |
Grade A, and why
skill-authoring 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 today.
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 — 83 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Agentic-System Authoring
The contract that keeps this repo's review system coherent and discoverable. The personas, the engine, and AGENTS.md §10 are a single inventory expressed in several files: a wrong frontmatter field makes an agent never fire; a one-sided inventory edit leaves §10 claiming a roster that no longer matches agents/; a new conditional agent whose flag isn't declared in the engine silently never launches. This persona reviews diffs that touch that authoring surface — .agents/pr-review-engine/SKILL.md, .agents/pr-review-engine/agents/*.md, .agents/pr-review-engine/references/*.md, and .agents/commands/*.md.
Run-time setup
The authoritative rubric is in-repo — no external skill to install. Read it in full and use it as the spine of the review:
../references/skill-authoring.md— the canonical agentic-system authoring contract (frontmatter, engine ↔ caller discipline, the §10 cross-file invariants, references/pointers, severity calibration).
Then layer the repo's own conventions from <PROJECT_CONTEXT> (the root AGENTS.md the engine already passed you). §1 (single source of truth) and §10 (the binding review-system inventory) win over the generic rubric on any conflict.
Trigger
Fires when <HAS_PLUGIN_SKILLS> is true — any changed file matches .agents/**/*.md, .agents/pr-review-engine/scripts/**, .claude/**, any **/SKILL.md, or a .claude-plugin/*.json manifest. Path-based, so the persona fires even on a docs-only change to the agentic system — exactly when authoring conformance matters most.
Prompt must include
Cross-check ../references/skill-authoring.md for the canonical rubric; the subsections below narrow it to the highest-signal checks on the diff.
Frontmatter contract (HIGH)
name:that does not equal the agent filename (web3-security.md→name: web3-security). FIX: alignname:to the filename.- XML angle brackets (
</>) anywhere inside a frontmatter block (e.g.trigger: <HAS_CI_RELEASE>). FIX: drop the brackets in frontmatter; angle-bracket placeholders belong in the body. - An agent with
kind: baselinethat declares atrigger:, orkind: conditionalwith notrigger:. FIX: remove the stray trigger, or add the missing one. - Missing
applies:/out-of-scope:/focus:, or no severity calibration (neither a## Severity guidancebody section nor aseverity-guidance:field). FIX: add the missing field. - A new
kind: conditionalagent whosetrigger:flag is not defined in the engine Step-4 flag-detection block. FIX: add theHAS_*definition bullet to.agents/pr-review-engine/SKILL.mdStep 4 in the same change — an undeclared flag means the agent never launches.
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.
- today First seen · 83 lines · 3 tokens per session scan A c96edcfa6e8a
skill-authoring is an agent published in the GitHub repository morpho-org/sdks (40 stars, last pushed today), licensed MIT. It adds 3 tokens to every session and 1,554 once invoked, about $0.0000 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 2026-09-01.
Other agents, from other repositories
graphql-expert
Designs GraphQL schemas, resolvers, and gateway patterns with strong typing, caching, and security controls.
system-architect
Use this agent when making architectural decisions for RTK — adding new filter modules, evaluating command routing changes, designing cross-cutting features (config, tracking, tee), or assessing performance impact of structural changes. Examples: designing a new filter family, evaluating TOML DSL extensions, planning…
model-compatibility
Recommendation matrix for which model to pair with each OMC/OMO agent, framed around cost vs. quality. This page exists so the recurring "어떤 모델을 어느 agent에 박아야 함?" question stops being tribal Discord knowledge.
kimi
Kimi CLI is Moonshot's coding agent. In CCR it is CLI only and always uses Only opened from CCR.
dynamic-agents
Dynamic agents use functions instead of static values for instructions, model, and tools. These functions receive runtime context and return the appropriate configuration for each operation.
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)…