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 mrinalwadhwa/fluent --skill review-behaviorsgit clone --depth 1 https://github.com/mrinalwadhwa/fluentWrote 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/mrinalwadhwa/fluent/review-behaviors)<a href="https://agentmods.dev/skills/mrinalwadhwa/fluent/review-behaviors"><img src="https://agentmods.dev/badge/skills/mrinalwadhwa/fluent/review-behaviors/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/mrinalwadhwa/fluent/review-behaviors"><img src="https://agentmods.dev/badge/skills/mrinalwadhwa/fluent/review-behaviors.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 | $0.00041 | $0.00340 |
| Opus 5 | $0.00020 | $0.00170 |
| Sonnet 5 | $0.00008 | $0.00068 |
| Haiku 4.5 | $0.00004 | $0.00034 |
Grade A, and why
review-behaviors 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 10d 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.
What it actually says
Purpose
Decide whether the behavior statements under review are fit to ship. Identify improvements that would make each statement more precise about what the system does, or the collection more coherent and complete as a whole.
Scope
The invoking layer decides what's in scope. For a diff-scoped review, that's the behavior statements changed in the diff. For a full-codebase audit, that's the entire set of behavior statements. Check each statement for quality and check the collection for coherence and completeness.
Method
-
Read
references/behaviors.mdfor behavior-writing standards. -
For each in-scope behavior statement:
- Verify it uses one of the six EARS patterns from
references/behaviors.md. Statements without a recognized trigger (WHEN, WHILE, IF, WHERE) or without "THE SYSTEM SHALL" are hard to verify. - Evaluate against the "Properties of a good behavior statement" and "What not to specify" sections in
references/behaviors.md. - Identify improvements.
- Verify it uses one of the six EARS patterns from
-
Check the collection for coherence — vocabulary consistency, contradictions, coverage, and redundancy — using the "Coherence across behavior statements" section of
references/behaviors.md. -
For each improvement, decide if it blocks shipping. Statements that leak implementation, combine multiple behaviors, or lack a way to verify typically block. Style issues and minor phrasing don't.
The invoking layer may add checks in addition to those above.
What ships with it
1 file beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
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.
- 10d ago First seen · 28 lines · 41 tokens per session scan A b72a4fb3e927
review-behaviors is a skill published in the GitHub repository mrinalwadhwa/fluent (88 stars, last pushed yesterday), licensed Apache-2.0. It adds 41 tokens to every session and 340 once invoked, about $0.0002 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
code_review
Code review — security audit, quality assessment, suggested fixes.
requesting-code-review
The formal protocol for requesting and processing peer code reviews from the Auditor sub-agent during CIEL execution.
coding-standards
Baseline cross-project engineering conventions for readability, immutability, and architecture.
everything-claude-code-conventions
Core repository standards for naming, structure, commits, and workflows.
ciel-review-and-adversarial
CIEL's adversarial review framework for rigorous code review, security auditing, and adversarial testing.
code-review
Revisão holística pré-commit (estilo PR Review). Valida o diff contra as 3 memórias vivas, a spec da task e Clean Code, com pré-condição de evidência de teste. Executada pelo Tech Lead; veredito escrito em review.md.