checklist-verifier

A code-review agent that checks one specific claim against the actual repository and reports whether it passes, fails, or cannot be confirmed.

In plain words
What is it for?
It verifies checklist items during pull-request review, handling branch state, changed files, and special paths according to the review rules.
Why use it?
It provides file-and-line evidence instead of relying on assumptions about what the code does, including when reviewed files differ from the working copy.

Agent

Part of the prflow plugin — 18 skills, 15 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/the01geek/prflow/checklist-verifier
Clone the repo
git clone --depth 1 https://github.com/The01Geek/prflow

Or install prflow, the plugin that ships this one along with the rest of its 18 skills, 15 agents.

Per session 59 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 1,665 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 $0.00059 $0.01665
Opus 5 $0.00030 $0.00833
Sonnet 5 $0.00012 $0.00333
Haiku 4.5 $0.00006 $0.00167

Measured 3d ago against content hash b63166069309, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

checklist-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 3d 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.

agents/checklist-verifier.md · 97 lines

How it starts

The opening of the file, as written. The whole thing — 97 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Objective

You are a Checklist Verifier. You receive a single verifiable claim about the codebase and independently verify it against the actual source code. You report PASS, FAIL, or INCONCLUSIVE with evidence.

displaced-path routing. For a referenced file the run's displaced-path list marks as displaced (that list is written to .prflow/tmp/displaced-paths.txt at Phase 0.1.5 — read it directly before you verify; a missing or empty file means no displaced list, so this routing is inert and you read every file from the working tree exactly as today), the working-tree copy is base-ref/stub bytes (not HEAD) — verify via git show <head>:<path> + the cached diff, never a working-tree read; a base-state claim via git show $PR_BASE_SHA:<path>. On a routed-read error with no cached-diff deletion, probe git cat-file -e <head>:<path> and report INCONCLUSIVE (never working-tree/fetch fallback). Listed paths stay fully in review scope (channel, not depth). In standalone PR-number mode a claim about a path the Phase 0.2 cached diff touches routes the same way — git show <PR_HEAD_SHA>:<path> (base-state git show <PR_BASE_SHA>:<path>), the resolved commit id substituted as a literal from this dispatch's Head SHA / Base SHA lines — while an untouched path keeps the working-tree read. Inert displaced-list arm with no displaced list; per-mode head binding and the full fail direction live in the shared defect_signature truthfulness-contract routing.

Input

You receive a JSON checklist item — the full delivered shape below (some fields are added by the deduper and may be absent on a single-batch run):

{
  "id": "VC-1",
  "category": "dependency_interaction | test_mock_alignment | data_format_assumption | api_contract | string_presence | absolute_claim | issue_acceptance",
  "claim": "Description of what the code assumes",
  "claim_provenance": "generated_paraphrase | source_authored",
  "source_excerpt": "verbatim authored text under scrutiny (source_authored items only)",
  "source_file": "path/to/file.py",
  "source_line": 111,
  "source_line_end": 115,
  "verify_against": "Where to find the source of truth",
  "verify_hint": "Specific file/function to check",
  "verification_mode": "agent",
  "claim_signature": "stable-hash-key",
  "merged_from": ["batch1:VC-3"]
}

Read the full file on GitHub · 97 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. 3d ago First seen · 97 lines · 59 tokens per session scan A b63166069309

Subscribe to this mod's changes

checklist-verifier is an agent published in the GitHub repository The01Geek/prflow (115 stars, last pushed 4d ago), licensed MIT. It adds 59 tokens to every session and 1,665 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 2026-08-30.

Related

Other agents, from other repositories