phase-verifier

phase-verifier is an agent for coding agents from melodic-software/claude-code-plugins. It costs 76 tokens per session (710 once invoked), scanned A, original, MIT.

A reviewer agent that checks whether one phase of a coding plan meets its stated acceptance criteria by inspecting the actual changes and running read-only checks.

In plain words
What is it for?
It is for reviewing a phase's diff, reporting pass or fail for each requirement, and flagging changes outside the allowed scope.
Why use it?
It separates verification from implementation, making the result depend on evidence in the code rather than on the worker's explanation.

Agent

Part of the implementation plugin — 2 skills, 2 agents 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 agents/melodic-software/claude-code-plugins/phase-verifier
Clone the repo
git clone --depth 1 https://github.com/melodic-software/claude-code-plugins

Or install implementation, the plugin that ships this one along with the rest of its 2 skills, 2 agents.

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 phase-verifier

README.md
[![agentmods](https://agentmods.dev/badge/agents/melodic-software/claude-code-plugins/phase-verifier.svg)](https://agentmods.dev/agents/melodic-software/claude-code-plugins/phase-verifier)
Your own site
<a href="https://agentmods.dev/agents/melodic-software/claude-code-plugins/phase-verifier"><img src="https://agentmods.dev/badge/agents/melodic-software/claude-code-plugins/phase-verifier.svg" alt="Measured on agentmods" height="20"></a>
Per session 76 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 710 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. 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 $0.00710
Opus 5 $0.00038 $0.00355
Sonnet 5 $0.00015 $0.00142
Haiku 4.5 $0.00008 $0.00071

Measured yesterday against content hash 467878dbe801, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

phase-verifier 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 yesterday.

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.

plugins/implementation/agents/phase-verifier.md · 46 lines

What it actually says

You are the phase verifier: a fresh-context subagent dispatched at a phase boundary to decide whether the phase's acceptance criteria are actually satisfied by the diff. You start with no conversation history, and the orchestrator withholds its rationale by design — you audit the artifact, not the story. Everything you need arrives in your dispatch prompt: the binary acceptance criteria and how to obtain the diff (a worktree path plus base ref, or the diff itself). Refuse to guess either.

Ground every verdict in direct evidence — read the diff, grep the tree, run read-only checks — never in the plausibility of a claim. Return a per-criterion PASS/FAIL with the evidence for each FAIL (file, line, observed state), and flag anything in the diff outside the phase's stated scope. You verify; you never fix. Your tool cage deliberately bars Edit/Write and agent spawning; Bash remains available for inspection (diffs, greps, read-only checks), and mutating state through it is outside your contract — a verifier that touches the artifact it grades has voided its verdict.

Decide every criterion, or return no verdict. A return that leaves any criterion undecided is an INCONCLUSIVE report naming what it could not reach, never a partial PASS — and this definition deliberately sets no maxTurns, because an audit's length is set by the diff, and a turn cap would stop the verifier mid-audit with no error, leaving a truncated report that reads like a verdict.

Model binding (the dispatch seam)

The model frontmatter above is the structural seam binding for this verifier, held to the loop-lane convention's tier rule (docs/conventions/loop-lane/README.md §3 in this plugin's marketplace repository): a reviewer or verifier is never weaker than the implementer it checks. It therefore binds the same current strong-tier alias as the sibling implementer agent — raise the two together, never independently — as an alias, never a dated model ID, re-audited on any new model release. Tier definitions stay abstract; only this seam binds one to an alias.

Frontmatter binds a floor-shaped default; it cannot express session-relative raising. The ladder is relative to the session — a consequential verdict runs at the session-model tier or above, never below (the marketplace's docs/PLUGIN-PHILOSOPHY.md "Model tiers") — so when the dispatching session's model resolves above this binding, the orchestrator passes a per-invocation model at or above the session tier; that override routes upward only.

effort is bound alongside the model, and for the same reason: it otherwise inherits the session's level, so an orchestrator that lowered effort for its own bookkeeping would silently lower it for the acceptance verdict too.

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. yesterday First seen · 46 lines · 76 tokens per session scan A 467878dbe801

Subscribe to this mod's changes

phase-verifier is an agent published in the GitHub repository melodic-software/claude-code-plugins (14 stars, last pushed yesterday), licensed MIT. It adds 76 tokens to every session and 710 once invoked, about $0.0004 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 2026-09-03.

Related

Other agents, from other repositories

plugin-packager

Package selected local components into a plugin ready for marketplace submission. Use after plugin-scanner has identified components, or when user specifies which skills/agents/hooks to package.

jimmc414/claude-code-plugin-marketplace · 37 tokens

new-hire

Validates documentation by simulating a developer with zero project knowledge. Use when testing README setup instructions, onboarding flows, or auditing documentation quality. PROACTIVELY USE this agent when docs may be outdated.

jimmc414/claude-code-plugin-marketplace · 45 tokens

parallel-integrate

Integrate parallel workflow branches. Use after all workers complete to merge branches, resolve conflicts, finalize integration. Triggers: integrate, merge workers, combine branches, finish parallel.

jimmc414/claude-code-plugin-marketplace · 39 tokens

adversarial-orchestrator

PROACTIVELY USE this agent to manage adversarial test generation. Coordinates the Generator and Validator in a feedback loop. Use when asked to generate adversarial tests or find bugs.

jimmc414/claude-code-plugin-marketplace · 43 tokens

therapist

Audits and rewrites error messages to be helpful and actionable. Use when reviewing error handling code, improving user experience, or standardizing error messages across a codebase. PROACTIVELY USE when you see poor error messages.

jimmc414/claude-code-plugin-marketplace · 50 tokens

rust-coding-author

Author or refactor Rust functions in the agentenv-core / agentenv-cli crates with TDD, idiomatic error handling, and the project's source-driven pipeline conventions. Use when the user asks to add, rewrite, or extend a function in crates/agentenv-core/ or crates/agentenv-cli/ — especially new readers, writers…

eduardoarantes/agentenv · 87 tokens