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.
npx agentmods add skills/agiflow/ai-plugin/review-worknpx skills add AgiFlow/ai-plugin --skill review-workgit clone --depth 1 https://github.com/AgiFlow/ai-pluginWrote 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.
[](https://agentmods.dev/skills/agiflow/ai-plugin/review-work)<a href="https://agentmods.dev/skills/agiflow/ai-plugin/review-work"><img src="https://agentmods.dev/badge/skills/agiflow/ai-plugin/review-work.svg" alt="Measured on agentmods" height="20"></a>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.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5 | $0.00072 | $0.01541 |
| Opus 5 | $0.00036 | $0.00771 |
| Sonnet 5 | $0.00014 | $0.00308 |
| Haiku 4.5 | $0.00007 | $0.00154 |
Grade A, and why
review-work 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 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.
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.
This is a copy
100% identical to review-work — 0 lines differ, which has more behind it and is treated as the original. This page carries a canonical link to it rather than competing with it.
How it starts
The opening of the file, as written. The whole thing — 184 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Invoked as
/agiflow:review-work. In hosts without slash-prompts, this skill is triggered by matching intent and drives AgiFlow via its MCP tools.
Usage:
/agiflow:review-work <work-unit-slug-or-id>- Review a specific work unit/agiflow:review-work <task-slug-or-id>- Review a specific task/agiflow:review-work- List completed items for review
Examples:
/agiflow:review-work DXX-WU-1(review work unit by slug)/agiflow:review-work DXX-3(review task by slug)/agiflow:review-work(interactive selection)
Purpose Verify that completed work actually meets its acceptance criteria, catches quality issues, and is ready to ship. AI code has 1.7x more defects than human-written code — review is the last line of defense before shipping.
Guardrails
- Review what exists — do not implement fixes during review (create follow-up tasks instead).
- Every acceptance criterion gets a pass/fail verdict with evidence.
- Flag issues by severity: blocker (must fix), warning (should fix), note (nice to fix).
- Be honest — a "pass" with gaps is worse than a clear "needs rework".
If a work unit or task slug/id is provided, load it with get_work_unit / get_task; otherwise list candidates with list_work_units / list_tasks for selection.
AgiFlow Project Management Guidelines
Follow the shared AgiFlow project-management guidelines in references/agiflow-agents.md — agent assignment, the task status workflow and transitions, work-unit best practices, and the tags strategy apply to this workflow.
Steps
1. Load the Work for Review
If work unit slug/id provided:
- Use
get_work_unitMCP tool to load the work unit and its tasks. - For each task in the work unit, use
get_taskto load full details (acceptance criteria, devInfo, comments).
If task slug/id provided: 3. Use get_task MCP tool to load the task with full details.
If nothing provided: 4. Use list_work_units with status: "completed" to show completed work units. 5. Use list_tasks with status: "Review" or status: "Done" to show completed tasks. 6. Ask user to select what to review.
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.
- 4d ago First seen · 184 lines · 72 tokens per session scan A 88f3a4a71bd6
review-work is a skill published in the GitHub repository AgiFlow/ai-plugin (2 stars, last pushed 1mo ago), licensed MIT. It adds 72 tokens to every session and 1,541 once invoked, about $0.0004 per session on Opus 5. A static security scan graded it A with 0 findings. It is 100% identical to review-work, differing in 0 lines, and is treated as a copy.
Other skills, from other repositories
review-work
Quality gate: verify each acceptance criterion of a completed task/work unit, run quality checks, and create follow-up tasks for gaps. Use before merging or to audit delivered work. Invoked as /agiflow:review-work . Uses getworkunit, gettask, updatetask, createtask, createtaskcomment.
skeptic-review
Use when you want an adversarial steelman pass on a text artifact before it ships — surface the load-bearing claims, the strongest unaddressed counter-position, the top failure modes, and any causality-vs-correlation gaps. Encodes the Skeptic deliberator role from the agent-council 5-perspective quality gate. Use…
vision-review
Provider-neutral rendered visual quality gate for UI/UX, game art and generated assets. Reviews deterministic constraints plus reference fidelity, hierarchy, readability, anatomy/material/lighting and technical artifacts against a validated Visual Basis.
design-critique
Give a structured product design critique — user job clarity, hierarchy, affordance, error states, accessibility, and consistency — focused on what to change, in what order, and why.
rubber-duck
Adversarial "rubber duck" review that turns explaining-out-loud into a hallucination check. The main session is the PRESENTER (it did the work — a design doc, investigation, or analysis — and holds the real reasoning) and reconstructs the topic to a LISTENER — a spawned subagent pinned to a DIFFERENT-vendor model that…
dispatch
Use when a task file exists in .hyperflow/tasks/ and workers need dispatching. Fans out parallel workers under per-batch Reviewers, runs a final integration review, and commits per sub-task. Endpoint of the auto-chain — no auto-deploy. Trigger with /hyperflow:dispatch, "run the plan", "execute the task", "build it"…