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 cruxible-ai/cruxible --skill review-threadgit clone --depth 1 https://github.com/cruxible-ai/cruxibleWrote 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/cruxible-ai/cruxible/review-thread)<a href="https://agentmods.dev/skills/cruxible-ai/cruxible/review-thread"><img src="https://agentmods.dev/badge/skills/cruxible-ai/cruxible/review-thread/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/cruxible-ai/cruxible/review-thread"><img src="https://agentmods.dev/badge/skills/cruxible-ai/cruxible/review-thread.svg" alt="Reviewed on agentmods" width="80" height="20"></a>- NVIDIA SkillSpector pass
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 | $0.00056 | $0.02554 |
| Opus 5 | $0.00028 | $0.01277 |
| Sonnet 5 | $0.00011 | $0.00511 |
| Haiku 4.5 | $0.00006 | $0.00255 |
Grade A, and why
review-thread 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 12d 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 — 213 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Review Thread (agent-operation)
Agent skill for recording review verdicts and maintaining the review-comment
thread on an agent-operation state instance. Read this before changing any
ReviewRequest.status or writing a review StateNote.
There is no core "review" verb. Everything here is built from the generic
graph primitives — batch-direct-write, entities, relationships — plus the
agent-operation kit's StateNote, ReviewRequest, and
state_note_about_review_request types. The kit enforces the discipline with a
single co_write mutation guard; this skill is how you satisfy it cleanly.
The one rule the guard enforces
The review_verdict_requires_rationale_note mutation guard rejects any write
that advances a ReviewRequest.status to changes_requested, approved, or
withdrawn unless the same write also creates a new
StateNote(kind=review_note) linked to that ReviewRequest via
state_note_about_review_request.
A verdict therefore cannot advance state without recording why. The note is the durable rationale; the status is just the headline.
This guard is always-on — it applies in every mode, governed or not. That is deliberate and is not a weaker form of governance: unlike an actor-credential guard (which needs a privileged identity to satisfy), this requirement is satisfiable in any mode by simply including the note in the batch. There is no credential to hold and no mode to be in; the cost of satisfying it is one extra entity + two edges, which this skill makes routine.
The guard only fires on a transition of status into a verdict value.
Re-asserting the same status (e.g. writing approved on an already-approved
request alongside an unrelated field edit) is not a transition and does not
fire. The non-verdict statuses (requested, in_review) are not guarded.
Recording a verdict — one atomic batch-direct-write
A verdict is one batch-direct-write unit of work, producing one
receipt. The single batch:
- Updates
ReviewRequest.statusto the verdict value. - Creates a
StateNote{kind: review_note, ...}carrying the rationale. - Creates the
state_note_about_review_requestedge (StateNote → ReviewRequest). - Creates the
state_note_authored_by_actoredge (StateNote → Actor).
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.
- 12d ago First seen · 213 lines · 56 tokens per session scan A 488ebff50433
review-thread is a skill published in the GitHub repository cruxible-ai/cruxible (17 stars, last pushed today), licensed Apache-2.0. It adds 56 tokens to every session and 2,554 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.
Other skills, from other repositories
trace-mcp-refactoring
Safe refactoring workflow using trace-mcp — assess risk, find candidates, check impact, and rename symbols across all files without missing import sites or cross-file references.
m1nd-first
Use when investigating a repository, searching for implementation, reviewing changes, working from specs/docs, or preparing a risky code change in an environment where m1nd is available. This doctrine makes m1nd the first investigative layer before grep, glob, or manual file reads, except when the task is pure…
audit-architecture
Audit module coupling, cohesion, hidden dependencies, and design smells. Use when conducting architecture reviews, evaluating design decisions, or identifying structural tech debt.
skill-scope-control
Analisa se alterações, diffs e PRs continuam alinhados ao objetivo declarado, identificando escopo extra e propondo manter, separar ou justificar.
find-bugs
Locate likely bug locations in code.
review-guidelines
Additional code review guidelines specific to this codebase. Auto-loaded by the review agent during PR reviews to enforce project conventions.