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 skills add atman-33/workhub --skill tree-reviewgit clone --depth 1 https://github.com/atman-33/workhubWrote 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/atman-33/workhub/tree-review)<a href="https://agentmods.dev/skills/atman-33/workhub/tree-review"><img src="https://agentmods.dev/badge/skills/atman-33/workhub/tree-review/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.
<a href="https://agentmods.dev/skills/atman-33/workhub/tree-review"><img src="https://agentmods.dev/badge/skills/atman-33/workhub/tree-review.svg" alt="Reviewed on agentmods" width="80" 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 5.1 | $0.00138 | $0.01033 |
| Opus 5.5 | $0.00055 | $0.00413 |
| Sonnet 5.5 | $0.00028 | $0.00207 |
| Haiku 4.5 | $0.00014 | $0.00103 |
Grade A, and why
tree-review 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 5d 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.
How it starts
The opening of the file, as written. The whole thing — 90 lines — stays where its author put it; the contents beside it link to each section on GitHub.
tree-review
The input is a tree that already exists: a problem at the root, causes below
it. ladder up builds such a tree from nothing by asking; this skill takes
the finished tree and does two jobs on it.
- review — inspect the logic. You answer with findings.
- group — boil the leaf causes down to a few major issues. You ask one question at a time and the user decides.
Take the mode from the argument. With none, run review first and offer group once the findings are dealt with. Read the tree in full before the first output: a file, a mind map note, or pasted text. Keep the user's wording for every node.
Review
Report findings only; do not edit the tree. Each finding names the node, quotes it, and says in one sentence what is wrong. Checks, in this order:
- Symptom mixed with cause. A node that restates its parent in other words is not a cause ("tasks pile up late" under "tasks increase"). Ask what produces it.
- Stopped at "not done". "We did not review X" is where a cause starts. Dig to why it was not done: no time, no owner, not seen as worth it.
- Mixed axes under one parent. Siblings should answer the same question. "Why is it slow now" and "why does it not get faster" are different questions and belong under different parents.
- Loops back to the root. A tree shows one direction, so a leaf that feeds the root hides a vicious circle. Name the loop explicitly: leaf → … → root → leaf.
- Fact or hypothesis. Mark each node as observed or inferred. List the inferred ones as hypotheses, with what would confirm them.
Close with a short list: findings by check, then the edits you suggest in
words ("move X under Y", "split Z", "add the cause under W"). When the tree
lives in a vault mind map, say the suggested edits can be handed to
workhub:diagram-edit; this skill never writes the file itself.
Group
Start from the leaves, after the review findings are fixed or set aside. Ask one question per turn, building each from the user's last answer.
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.
- 5d ago Changed 676c293cda96
- 9d ago First seen · 90 lines · 138 tokens per session scan A e037151e8d14
tree-review is a skill published in the GitHub repository atman-33/workhub (3 stars, last pushed yesterday), licensed MIT. It adds 138 tokens to every session and 1,033 once invoked, about $0.0006 per session on Opus 5.5. 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 2026-10-07.
Other skills, from other repositories
gonavi-cli
Operate databases through the GoNavi headless CLI — the gonavi executable shipped in verified GitHub Release archives. Covers listing/adding/importing saved connections, running SQL queries against saved connections or ad-hoc connection files, exporting result sets to csv/json/md/html/xlsx, batch-executing SQL files…
trellis-brainstorm
Guides collaborative requirements discovery before implementation. Creates task directory, seeds PRD, asks high-value questions one at a time, researches technical choices, and converges on MVP scope. Use when requirements are unclear, there are multiple valid approaches, or the user describes a new feature or complex…
readmd-skill-creator
Use when a ReadMD user wants to design, test, or improve a reusable document-writing Skill.
trellis-break-loop
Deep bug analysis to break the fix-forget-repeat cycle. Analyzes root cause category, why fixes failed, prevention mechanisms, and captures knowledge into specs. Use after fixing a bug to prevent the same class of bugs.
avoid-ai-writing
Use when outward-facing Markdown should avoid recognizable machine-written patterns while retaining its claims.
creative-writing-craft
Use when fiction or creative Markdown needs stronger character, scene, voice, or narrative craft.