check-first

check-first is a skill for Claude Code, Codex from bearded-illirian/trailmark. It costs 98 tokens per session (1,107 once invoked), scanned A, original, MIT.

A requirements-checking step that compares a source document with proposed work units such as blocks, epics, or tasks.

In plain words
What is it for?
Use it after breaking down a brief, specification, or product document to produce a coverage table and stop when gaps need a decision.
Why use it?
It finds requirements that were missed before work receives final approval.

Skill for Claude CodeCodex

Written for no agent in particular: nothing here depends on one. Also seen: installed under .agents/ (shared by several agents); $skill-name invocation.

Good fit Use it after breaking down a brief, specification, or product document to produce a coverage table and stop when gaps need a decision.

Compare 6 skills from other repositories ↓
Install with agentmods
npx agentmods add skills/bearded-illirian/trailmark/check-first
Install

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.

Any agent
npx skills add bearded-illirian/trailmark --skill check-first
Clone the repo
git clone --depth 1 https://github.com/bearded-illirian/trailmark

Made for: Claude Code, Codex.

Wrote 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.

agentmods badge for check-first

README.md
[![agentmods](https://agentmods.dev/badge/skills/bearded-illirian/trailmark/check-first.svg)](https://agentmods.dev/skills/bearded-illirian/trailmark/check-first)
Your own site
<a href="https://agentmods.dev/skills/bearded-illirian/trailmark/check-first"><img src="https://agentmods.dev/badge/skills/bearded-illirian/trailmark/check-first.svg" alt="Measured on agentmods" height="20"></a>
Per session 98 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 1,107 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. A grade says what 26 rules found in the file — not that it is safe. Third-party audits
  • NVIDIA SkillSpector pass 7 Sept 2026
How audits are shown
Origin original No closer match found in the catalogue.
Token cost

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.

ModelPer sessionOnce invoked
Fable 5.1 $0.00098 $0.01107
Opus 5 $0.00049 $0.00553
Sonnet 5 $0.00020 $0.00221
Haiku 4.5 $0.00010 $0.00111

Measured 4d ago against content hash f430f24ee12d, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-08, from the pricing page.

Security

Grade A, and why

check-first 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 4d 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.

adapters/codex/.agents/skills/check-first/SKILL.md · 157 lines

How it starts

The opening of the file, as written. The whole thing — 157 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Check-First Protocol

Runs after a decomposition step in any workflow (e.g. after proposing epics for a brief, or blocks for a specification). Verifies that every requirement from the source document is covered by at least one proposed unit.


Step 1 — Extract requirements

From the already-read requirements document (brief, specification, technical description, product doc) extract every requirement. Look in sections like:

  • Scope / What's included
  • Acceptance criteria
  • What must exist
  • Functional requirements
  • Non-functional requirements

Each requirement is one row. Don't duplicate near-identical items — merge them if the essence is the same.


Step 2 — Output a coverage table

Requirements coverage check:

┌─────┬────────────────────────────────────┬──────────────┬────────────┐
│  #  │ Requirement                        │ Unit         │ Status     │
├─────┼────────────────────────────────────┼──────────────┼────────────┤
│ 1   │ [requirement from source doc]      │ Unit N       │ ✅ Covered │
│ 2   │ [requirement from source doc]      │ Unit N       │ ✅ Covered │
│ 3   │ [requirement from source doc]      │ —            │ ⚠️ Gap     │
└─────┴────────────────────────────────────┴──────────────┴────────────┘

"Unit" = the smallest labelled thing in your decomposition: an epic, a block, a task — whatever the current workflow produces.


Step 3 — Decision

If every row is ✅:

✅ All requirements covered — continuing.

Proceed without pause.

If there are ⚠️ gaps:

⚠️ N gap(s) found. Options:
1. Add to an existing unit [Unit X]
2. Create a new unit
3. Move to backlog — not part of this cycle

Wait for user decision per gap. After decisions land — update the unit list and continue.


Rules

  • Don't skip even when the decomposition looks obviously complete
  • On gaps — don't proceed without a user decision
  • Extract requirements from the source text, don't invent them
  • The point of the protocol is a paper trail: even a "no gaps" result is a first-class artifact — future readers see what was checked

Read the full file on GitHub · 157 lines

Files

What ships with it

1 file beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.

Changes

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.

  1. 4d ago Changed f430f24ee12d
  2. 8d ago First seen · 157 lines · 98 tokens per session scan A de34f34ad841

Subscribe to this mod's changes

check-first is a skill published in the GitHub repository bearded-illirian/trailmark (19 stars, last pushed 6d ago), licensed MIT. It adds 98 tokens to every session and 1,107 once invoked, about $0.0005 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 2026-08-30.