constatar-verify

constatar-verify is a skill for Claude Code from JairoTorregrosa/jaiskills. It costs 55 tokens per session (992 once invoked), scanned A, original, MIT.

A skill for running and interpreting constatar, a verification engine that checks coding work against evidence. Its checks range from commands such as tests and builds to repeated, held-out, or adversarial checks.

In plain words
What is it for?
Use it to run a constatar plan, inspect its JSON results, and audit whether a change meets the required checks.
Why use it?
It prevents work from being accepted based only on an agent's claim by requiring concrete evidence and applying a defined verification ladder.

Skill for Claude Code

Written for Claude Code: ${CLAUDE_PLUGIN_ROOT} variable. Also seen: mentions Codex.

Runs only inside its plugin — its command needs a path that Claude Code sets for a plugin’s own hooks and for nothing else. Install the plugin, not this.

Part of the jaiskills plugin — 12 skills, 13 commands, 4 agents, 2 hooks, 1 MCP server shipped together

Good fit Use it to run a constatar plan, inspect its JSON results, and audit whether a change meets the required checks.

Compare 6 skills from other repositories ↓
Install

Getting it into your agent

This one installs as part of its plugin. Adding the marketplace and installing the plugin brings it with everything else the plugin ships.

Claude Code
/plugin marketplace add JairoTorregrosa/jaiskills
Claude Code
/plugin install jaiskills

Made for: Claude Code.

Or install jaiskills, the plugin that ships this one along with the rest of its 12 skills, 13 commands, 4 agents, 2 hooks, 1 MCP server.

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 constatar-verify

README.md
[![agentmods](https://agentmods.dev/badge/skills/jairotorregrosa/jaiskills/constatar-verify/github.svg)](https://agentmods.dev/skills/jairotorregrosa/jaiskills/constatar-verify)
Your own site
<a href="https://agentmods.dev/skills/jairotorregrosa/jaiskills/constatar-verify"><img src="https://agentmods.dev/badge/skills/jairotorregrosa/jaiskills/constatar-verify/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.

agentmods 80×15 button for constatar-verify

Your own site · 80×15
<a href="https://agentmods.dev/skills/jairotorregrosa/jaiskills/constatar-verify"><img src="https://agentmods.dev/badge/skills/jairotorregrosa/jaiskills/constatar-verify.svg" alt="Reviewed on agentmods" width="80" height="20"></a>
Per session 55 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 992 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.
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 $0.00055 $0.00992
Opus 5 $0.00028 $0.00496
Sonnet 5 $0.00011 $0.00198
Haiku 4.5 $0.00006 $0.00099

Measured 11d ago against content hash f2b38ca9b7db, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-10, from the pricing page.

Security

Grade A, and why

constatar-verify 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 11d 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.

skills/orchestration/constatar-verify/SKILL.md · 74 lines

How it starts

The opening of the file, as written. The whole thing — 74 lines — stays where its author put it; the contents beside it link to each section on GitHub.

constatar: verification-first orchestration

constatar is a compiled Rust engine (constatar binary) that drives headless coding agents (claude, codex) and refuses to accept work without grounded evidence. You are the thin interface: read plans, invoke the binary, interpret its JSON. The engine owns all state (journal + git worktrees) and enforces all invariants in compiled code — never re-implement them in prompt logic, never bypass them.

Ladder reference (what each of the 6 rungs demands and produces): ${CLAUDE_PLUGIN_ROOT}/references/constatar/ladder.md. Plan schema: ${CLAUDE_PLUGIN_ROOT}/references/constatar/plan-schema.md.

The verification ladder (lowest rung = strongest = preferred)

  1. Deterministic oracle — commands whose exit codes are the evidence (tests, clippy, build).
  2. Property/metamorphic — invariant checks repeated N times (property_runs) to catch flake.
  3. Held-out — checks the executor never saw; pass requires ALL held-out checks passing AND hacking gap (visible rate − held-out rate) within threshold.
  4. Adversarial refutation — a cross-family skeptic whose default verdict is REFUTED; its grounding commands are executed by the engine, never trusted.
  5. Cross-provider judge — a different model family judges the unit's actual commit diff (engine-derived and hashed), hunting hacking behaviors.
  6. Peer review — last resort, only when 1–5 do not apply; grounding commands must touch the changed files.

Declared rungs form a conjunction: every declared rung must pass; the first failure is terminal (no falling through to weaker lenses). Prefer declaring the lowest rungs that genuinely verify the work; add rung 4 or 5 for units whose correctness a test suite cannot fully pin down.

Two invariants make a rung real, both engine-enforced:

  • Independence: verifiers at rungs 4–6 come from a different model family than the producer; their prompts never contain producer chain-of-thought.
  • Groundedness: evidence is executed commands with captured exit codes, artifact hashes of real git objects, or red-first observations. A verdict without grounding is unrepresentable.

Read the full file on GitHub · 74 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. 11d ago First seen · 74 lines · 55 tokens per session scan A f2b38ca9b7db

Subscribe to this mod's changes

constatar-verify is a skill published in the GitHub repository JairoTorregrosa/jaiskills (5 stars, last pushed 10d ago), licensed MIT. It adds 55 tokens to every session and 992 once invoked, about $0.0003 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 2026-08-31.

Related

Other skills, from other repositories

foundry-hosted-agent-validation

Step-by-step process for validating a Python Foundry hosted agent sample (under python/samples/04-hosting/foundry-hosted-agents/) end to end — running it locally (native runtime and azd ai agent run) and after deploying it to an Azure AI Foundry project with azd. Use this when asked to validate a hosted agent sample.

microsoft/agent-framework · 82 tokens

build-and-test

How to build and test .NET projects in the Agent Framework repository. Use this when verifying or testing changes.

microsoft/agent-framework · 26 tokens

verify-dotnet-samples

How to build, run and verify the .NET sample projects in the Agent Framework repository. Use this when a user wants to verify that the samples still function as expected.

microsoft/agent-framework · 40 tokens

regex-tester

Validate, test, and debug regular expressions by executing them against sample inputs. Use when asked to build, verify, or explain a regex pattern.

microsoft/agent-framework · 33 tokens

story-origin-check

Recover the first public timestamp and canonical major coverage for a newsjacking signal, then decide whether newer coverage is the same story, a different story, or a materially new development.

elvisun/newsjack · 40 tokens

prompt-proximity-architecture

Turn an approved measurement charter, ICPs, and buyer jobs into a budget-aware prompt coverage blueprint across proximity bands, aided status, information acts, journey states, roles, locales, evidence grades, partitions, and measurement lanes. Use before prompt wording to define required, optional, and prohibited…

elvisun/newsjack · 67 tokens