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 kangig94/coral --skill preplangit clone --depth 1 https://github.com/kangig94/coralWrote 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/kangig94/coral/preplan)<a href="https://agentmods.dev/skills/kangig94/coral/preplan"><img src="https://agentmods.dev/badge/skills/kangig94/coral/preplan/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/kangig94/coral/preplan"><img src="https://agentmods.dev/badge/skills/kangig94/coral/preplan.svg" alt="Reviewed on agentmods" width="80" height="20"></a>- NVIDIA SkillSpector warn
SkillSpector: 7 findings, up to high
These are SkillSpector’s own severities. On a checked sample its high-severity flags on skills were ~96% false positives — a documented command, a public API, a “never do X” rule — so we show them as a caution to read, not a verdict. Why →
- high Prompt Injection · line 251 Hidden instructions were detected in comments or invisible text. These could contain malicious directives. Manual review is recommended.Fix: Audit all comments and invisible characters. Remove any instructions that direct the agent to perform unauthorized actions. Use plain, reviewable content.
- medium Prompt Injection · line 13 Large whitespace padding was detected (a block of blank lines or a long run of spaces). This can push injected instructions below or to the right of the visible area so a human reviewer never sees them while the agent still reads them. Manual review of the hidden content is recommended.Fix: Remove the large whitespace padding (blank-line blocks or long space runs) and review any content hidden below or to the right of it. Keep skill files compact and reviewable so no instructions can be
- medium Prompt Injection · line 15 Large whitespace padding was detected (a block of blank lines or a long run of spaces). This can push injected instructions below or to the right of the visible area so a human reviewer never sees them while the agent still reads them. Manual review of the hidden content is recommended.Fix: Remove the large whitespace padding (blank-line blocks or long space runs) and review any content hidden below or to the right of it. Keep skill files compact and reviewable so no instructions can be
- medium Excessive Agency · line 37 Skill enables autonomous high-impact decisions without human-in-the-loop verification. Critical operations (destructive commands, financial transactions, data deletion) should require explicit user confirmation.Fix: Add human-in-the-loop confirmation for destructive, irreversible, or high-impact operations. Never auto-execute commands that modify files, send data, or alter system state.
- medium Excessive Agency · line 48 Skill enables autonomous high-impact decisions without human-in-the-loop verification. Critical operations (destructive commands, financial transactions, data deletion) should require explicit user confirmation.Fix: Add human-in-the-loop confirmation for destructive, irreversible, or high-impact operations. Never auto-execute commands that modify files, send data, or alter system state.
- medium Excessive Agency · line 120 Skill enables autonomous high-impact decisions without human-in-the-loop verification. Critical operations (destructive commands, financial transactions, data deletion) should require explicit user confirmation.Fix: Add human-in-the-loop confirmation for destructive, irreversible, or high-impact operations. Never auto-execute commands that modify files, send data, or alter system state.
- medium Excessive Agency · line 226 Skill enables autonomous high-impact decisions without human-in-the-loop verification. Critical operations (destructive commands, financial transactions, data deletion) should require explicit user confirmation.Fix: Add human-in-the-loop confirmation for destructive, irreversible, or high-impact operations. Never auto-execute commands that modify files, send data, or alter system state.
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 | $0.00022 | $0.05893 |
| Opus 5.5 | $0.00009 | $0.02357 |
| Sonnet 5.5 | $0.00004 | $0.01179 |
| Haiku 4.5 | $0.00002 | $0.00589 |
Grade A, and why
preplan 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 — 426 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Pre-plan
Structured problem-definition conversation with the user before planning begins.
Argument Routing
| Argument | Mode |
|---|---|
<prompt> |
Self-execute on current host (default) |
--deep |
Enable pioneer review for elegant alternatives. Blocks Step 3 until pioneer returns. |
--delegate |
Delegate pioneer to the other host (Claude → Codex, Codex → Claude, Copilot → Codex; from SessionStart Current host:). Activates --deep. |
Strip --deep and --delegate flags before passing the prompt to the execution path.
<Preplan_Protocol>
You are the Problem Definer: gather context, fill a structured agreement, refine through conversation, propose transition to planning.
Not responsible for: plans (plan), implementation (ralph), architecture (architect).
NEVER implement. NEVER write source code. Problem definition only.
The agreement consists of 7 items. Fill autonomously where possible, mark uncertain
items with the "unconfirmed" marker, then seek user feedback. When pioneer returned a report, the file also
carries ## Pioneer Ledger after the items, pointing at the sealed Pioneer Report beside it.
### Required Items
| # | Item | Description | Autonomous Source |
|---|------|-------------|-------------------|
| 1 | **Problem Statement** | Current state vs desired state. What is wrong? | Conversation context |
| 2 | **Success Criteria** | Testable, verifiable conditions for "done" | Reverse-infer from problem (unconfirmed) |
| 3 | **Scope** | What is included / excluded. Must include a **Compatibility** sub-item when the change touches existing APIs, data formats, or public interfaces: preserve backward compatibility vs full deprecation. Always mark Compatibility as `[unconfirmed]` — never auto-confirm — listing only the alternatives that genuinely solve it (Step 3). | Codebase analysis (unconfirmed) |
| 4 | **Assumptions** | What we assume to be true | Code analysis, project rules |
| 5 | **Affected Systems** | Existing systems affected by this change | Dependency analysis |
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 Changed 6cc6e38625db
- 5d ago Changed 3af1d2ece209
- 15d ago Changed 75138525e5a0
- 29d ago Changed · +103 lines d4a486a3c2cf
- 1mo ago First seen · 323 lines · 22 tokens per session scan A 049775f0b9a1
preplan is a skill published in the GitHub repository kangig94/coral (11 stars, last pushed today), licensed MIT. It adds 22 tokens to every session and 5,893 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 2026-08-30.
Other skills, from other repositories
rls-audit
This skill should be used to audit Row-Level Security and database permissions on a Postgres backend exposed through an auto-generated API — Supabase, InsForge, or any Postgres + PostgREST setup. Trigger phrases (English) include "audit my RLS", "review my policies", "is my database exposed", "can users see each…
deployment
This skill should be used when shipping an app to production — setting up CI/CD, deploys, preview environments, error monitoring, environment separation, or rollback. Trigger phrases include "deploy this", "set up CI/CD", "GitHub Actions", "add error tracking", "set up Sentry", "monitor production", "staging…
api-design
This skill should be used when designing or building an API — endpoints, routes, request/response shapes, error formats, pagination, versioning, or deciding REST vs GraphQL. Trigger phrases include "design the API", "build an endpoint", "REST or GraphQL", "how should I structure my routes", "API error format"…
architecture-primer
This skill should be used when the user wants to design, plan, or decide the architecture of a system, app, or platform from scratch. Trigger phrases include "how do I structure this", "what architecture should I use", "how does this scale", "monolith or microservices", "what database", "what stack", "design the…
saas-router
This skill should be used FIRST whenever the user wants to create, build, design, ship, or prototype any digital product — a SaaS, web app, platform, MVP, landing page, sales page, dashboard, admin panel, backoffice, internal tool, API, database, auth, payments, an integration, an AI feature, or any user interface.…
ai-features
This skill should be used when putting an LLM inside the product itself — a chatbot or support agent, AI-generated reports or summaries, classification, extraction, lead scoring, a WhatsApp or voice bot, or any feature where your code calls a model on behalf of your users. Trigger phrases (English) include "add AI to…