c-validate

c-validate is a skill for Claude Code, Codex from sentasity/cadence. It costs 76 tokens per session (1,519 once invoked), scanned A, original, MIT.

A workflow for checking a software plan after deployment, meaning after the new code has been released to its running environment.

In plain words
What is it for?
Use it to validate an implemented plan, run automated tests, guide manual browser checks, and mark the plan completed after a full pass.
Why use it?
It verifies prerequisites, automated checks, and user-visible behavior in a fixed order, and records which checks passed.

Skill for Claude CodeCodex

Part of the cadence plugin — 11 skills, 5 agents, 1 hook shipped together

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.

agentmods
npx agentmods add skills/sentasity/cadence/c-validate
Any agent
npx skills add sentasity/cadence --skill c-validate
Clone the repo
git clone --depth 1 https://github.com/sentasity/cadence

Made for: Claude Code, Codex.

Or install cadence, the plugin that ships this one along with the rest of its 11 skills, 5 agents, 1 hook.

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 c-validate

README.md
[![agentmods](https://agentmods.dev/badge/skills/sentasity/cadence/c-validate.svg)](https://agentmods.dev/skills/sentasity/cadence/c-validate)
Your own site
<a href="https://agentmods.dev/skills/sentasity/cadence/c-validate"><img src="https://agentmods.dev/badge/skills/sentasity/cadence/c-validate.svg" alt="Measured on agentmods" height="20"></a>
Per session 76 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 1,519 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 1 finding. Scan, not verified.
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 $0.00076 $0.01519
Opus 5 $0.00038 $0.00759
Sonnet 5 $0.00015 $0.00304
Haiku 4.5 $0.00008 $0.00152

Measured 4d ago against content hash 006fdffa29df, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

c-validate scanned grade A with 1 finding 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.

Makes network callslowCapability

Not a fault in itself. Listed so you know the mod talks to something, and to what.

**2. Category A — Automated.** Run each item itself: `curl`, `psql`, `pytest`, etc. Mark `- [x]` as each passes. Stop on first failure; surface exact output for user resolution.
skills/c-validate/SKILL.md · 77 lines

How it starts

The opening of the file, as written. The whole thing — 77 lines — stays where its author put it; the contents beside it link to each section on GitHub.

/c-validate

You walk a plan's post-deploy validation doc. You do NOT deploy. You verify deployed behavior by running automated checks, asking the user to do manual UI work, and checking observable signals.

Invocation

/c-validate <path-to-plan-folder>

Status gates

Plan status Behavior
draft / in-progress Refuse: "Plan isn't implemented yet. Run /c-execute first."
implemented Run the walk.
completed Run anyway (re-validation after a re-deploy). Reset checkboxes per config.
on-hold / superseded Refuse with status-mismatch error.

Walk order (strict)

Recommendation note (once, at the start of the walk). Per skills/_shared/browser-validation.md: if this plan's 96-validation.md has at least one Category B item with an e2e: reference AND no runner is configured or detected (validate.browser_command left at default npx playwright test and auto detection finds no suite), print one informational line recommending a runner (Playwright by default), then proceed — those items take the manual fallback. A repo with a runner configured/detected prints nothing. Once per run; never nag.

1. Category C — Prerequisites first. Read the plan's 96-validation once via skills/_shared/storage-resolution.md (read_artifact); for section C print the prereq list, ask user to confirm each: "Backend deployed? Migration run? Test users seeded? (y/each)" Walk BLOCKS until every C item confirmed. No prereq = no validation.

2. Category A — Automated. Run each item itself: curl, psql, pytest, etc. Mark - [x] as each passes. Stop on first failure; surface exact output for user resolution.

3. Category B — Manual workflow last. Resolve validate.browser_driver to delegate-or-manual once at the start of this pass, per skills/_shared/browser-validation.md (auto runs suite detection and delegates if found; playwright forces delegation; manual forces manual). Then, for each Category B item:

  • If the driver resolved to delegate AND the item carries an e2e: line → delegate. Compose <browser_env_preamble> && <browser_command> <spec> -g "<grep>" (drop the && prefix if no preamble; drop -g if the item has none), run it headless and non-interactively, and map the exit code: exit 0 → mark - [x]; non-zero → a validation failure, routed through the existing Failure handling section below (stop the walk, surface exact output, offer Fix / OOS / Abort). No new failure path.
  • Otherwise → manual (today's behavior). Print the human-readable step list ("Log in as X → click Y → see Z"); user does the clicks; run the verification (DB query or API call) after each step or at walkthrough's end; mark - [x] when verified.

Read the full file on GitHub · 77 lines

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 First seen · 77 lines · 76 tokens per session scan A 006fdffa29df

Subscribe to this mod's changes

c-validate is a skill published in the GitHub repository sentasity/cadence (1 stars, last pushed 9d ago), licensed MIT. It adds 76 tokens to every session and 1,519 once invoked, about $0.0004 per session on Opus 5. A static security scan graded it A with 1 finding (makes network calls). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.