review-queue

A command for sorting items waiting for a person's decision or action in a review queue. The queue groups items such as decisions, follow-up actions, testing, deferred visual issues, and removable clutter.

In plain words
What is it for?
Use it to review all pending items or focus on one group, then refine discovered issues into tasks or record the required decision.
Why use it?
It keeps human judgments and unfinished checks from getting lost among automated sprint work.

Command

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 commands/kesteva/soloflow/review-queue
Clone the repo
git clone --depth 1 https://github.com/kesteva/soloflow
Per session 28 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 8,669 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 $0.00028 $0.08669
Opus 5 $0.00014 $0.04334
Sonnet 5 $0.00006 $0.01734
Haiku 4.5 $0.00003 $0.00867

Measured 2d ago against content hash 778c39811c37, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

review-queue 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 2d 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.

The Testing bucket holds *all* human verification work — visual flows (Maestro / Playwright) AND non-visual manual checks ("curl /api/foo and confirm 200", "open the page in Safari and confirm copy"). Both are handled he
commands/review-queue.md · 695 lines

How it starts

The opening of the file, as written. The whole thing — 695 lines — stays where its author put it; the contents beside it link to each section on GitHub.

/soloflow:review-queue

Standalone triage for .soloflow/human-review-queue.md. Runs outside /soloflow:sprint's end-of-sprint flow and can be invoked any time. The interactive phases (cruft → decisions → actions → testing → deferred visual) are fast and decision-only; all time-intensive agent work (re-verification, issue refinement) runs in a single end-of-run batch so you can walk away once the quick decisions are made.

$ARGUMENTS optionally includes one bucket-scope flag:

  • --skip-cruft — skip Step 1 (useful for quick re-invocations after a prior cruft pass).
  • --decisions-only — handle the Decisions bucket only.
  • --actions-only — handle the Actions bucket only.
  • --testing-only — handle the Testing bucket only (visual + non-visual manual verification).
  • --deferred-visual-only — handle only the Deferred Visual bucket (visual failures the user previously chose to defer).

If $ARGUMENTS is non-empty and not one of the recognized flags, print usage and stop.

The bucket model:

  • Decisions — judgment calls (UX/copy/scope/security tradeoffs). The human reads context and picks. No re-verification path.
  • Actions — operational work the human performs on systems (deploy, configure, set env vars, run migrations, resolve merge conflicts, install tooling). After completion, the verifier can re-run the previously-blocked check.
  • Testing — verification only a human can do (visual checks, manual flows, ground-truth checks like "hit /api/foo and confirm 200"). The human's confirmation IS the verification — no agent re-verifies.
  • Deferred Visual — visual verification failures the user explicitly chose to defer in a previous run; holding area before promoting to a backlog TASK or re-testing.

Model resolution (applies to every Agent spawn below)

Before invoking the Agent tool, resolve models.<name> per the three-tier recipe in docs/CUSTOMIZATION.md#config-resolution and pass the resolved value as the Agent tool's model parameter.

Read the full file on GitHub · 695 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. 2d ago First seen · 695 lines · 28 tokens per session scan A 778c39811c37

Subscribe to this mod's changes

review-queue is a command published in the GitHub repository kesteva/soloflow (41 stars, last pushed 3mo ago), licensed MIT. It adds 28 tokens to every session and 8,669 once invoked, about $0.0001 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 2026-08-30.