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 figitaki/reviews --skill writing-review-packetsgit clone --depth 1 https://github.com/figitaki/reviewsWrote 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/figitaki/reviews/writing-review-packets)<a href="https://agentmods.dev/skills/figitaki/reviews/writing-review-packets"><img src="https://agentmods.dev/badge/skills/figitaki/reviews/writing-review-packets/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/figitaki/reviews/writing-review-packets"><img src="https://agentmods.dev/badge/skills/figitaki/reviews/writing-review-packets.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 | $0.00082 | $0.01498 |
| Opus 5 | $0.00041 | $0.00749 |
| Sonnet 5 | $0.00016 | $0.00300 |
| Haiku 4.5 | $0.00008 | $0.00150 |
Grade A, and why
writing-review-packets 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 — 143 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Writing Review Packets
Use this skill when drafting packet markdown for reviews push --packet. A good packet is a reviewer map: it groups related hunks, explains why to look, and avoids rephrasing the diff line-by-line.
Packet Markdown Format
Use one top-level title and ## packet sections:
# Packet Title
Short overview of what the review packet covers.
## Section Title
One or two sentences orienting the reviewer.
One sentence of context before the hunk when useful.
@hunk path/to/file.ex#1
Rules:
- Start with exactly one
#title. - Use
##for review packet sections. - Prose between hunk refs is allowed and encouraged.
- Hunk refs use
@hunk path#N. - Paths and hunk numbers must match the diff being pushed.
Writing Method
- Inspect the diff with
git diffor the intended CLI range. - Identify logical review areas: persistence/schema, read model, LiveView state, rendering, styling, tests, tooling.
- Prefer stable section titles if updating an existing packet and you want approvals to inherit.
- Put a 1-2 sentence summary at the top of each
##section. - Use
###subheadings sparingly, only when a section truly needs scan landmarks; do not add a stock technical-overview subsection to every section. - Add terse hunk explainers before hunks when they help the reader know what to look for.
- Avoid hunk explainers that restate the diff; explain context, dependency order, risk, or review intent.
- Keep section decisions independent from hunk viewed progress in wording.
Structure Preferences
Treat each ## section as an approval unit. A reviewer should be able to approve the section without implicitly approving unrelated risks.
Shape sections around independent review decisions, not just file adjacency. Split a section when it mixes substantial concerns such as:
- persistence/schema changes and runtime behavior,
- backend API contracts and CLI/client wiring,
- UI behavior/templates and design-system or CSS primitives,
- implementation hunks and broad regression coverage,
- feature work and packet/tooling/documentation changes.
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.
- 12d ago First seen · 143 lines · 82 tokens per session scan A 204a484bd393
writing-review-packets is a skill published in the GitHub repository figitaki/reviews (6 stars, last pushed 7d ago), licensed MIT. It adds 82 tokens to every session and 1,498 once invoked, about $0.0004 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 2026-08-31.
Other skills, from other repositories
elixir-phoenix-review
Review Phoenix changes with parallel specialists before commit or PR.
elixir-phoenix-pr-review
Handle Phoenix PR review comments: inspect feedback, draft replies, and fix code.
elixir-phoenix-triage
Triage review findings after $elixir-phoenix-review before fixing.
elixir-phoenix-runtime-durability-review
Elixir/Phoenix: Review lifecycle, state-machine, Oban, persistence, pause/resume, retry, and restart-sensitive changes before commit, push, or PR. Use for concurrency-sensitive runtime work to produce explicit blocking vs optional findings, require durability checks, and verify smoke plus restart resilience when…
elixir-phoenix-techdebt
Assess Phoenix technical debt, duplication, cleanup, and refactor targets.
elixir-review
Review Elixir/Phoenix code: OTP patterns, GenServer, LiveView, fault tolerance and concurrency.