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 tomzx/agents --skill review-pr-fullgit clone --depth 1 https://github.com/tomzx/agentsWrote 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/tomzx/agents/review-pr-full)<a href="https://agentmods.dev/skills/tomzx/agents/review-pr-full"><img src="https://agentmods.dev/badge/skills/tomzx/agents/review-pr-full/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/tomzx/agents/review-pr-full"><img src="https://agentmods.dev/badge/skills/tomzx/agents/review-pr-full.svg" alt="Reviewed on agentmods" width="80" height="20"></a>- NVIDIA SkillSpector warn
SkillSpector: 1 finding, up to medium
These are SkillSpector’s own severities. On a checked sample its high-severity flags on skills were ~96% false positives — a documented command, a public API, a “never do X” rule — so we show them as a caution to read, not a verdict. Why →
- medium Excessive Agency · line 230 Skill enables autonomous high-impact decisions without human-in-the-loop verification. Critical operations (destructive commands, financial transactions, data deletion) should require explicit user confirmation.Fix: Add human-in-the-loop confirmation for destructive, irreversible, or high-impact operations. Never auto-execute commands that modify files, send data, or alter system state.
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 | $0.00108 | $0.10115 |
| Opus 5.5 | $0.00043 | $0.04046 |
| Sonnet 5.5 | $0.00022 | $0.02023 |
| Haiku 4.5 | $0.00011 | $0.01012 |
Grade A, and why
review-pr-full 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 today.
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 — 627 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Review PR Full
Runs the complete review pipeline on a single PR: /assess-pr-risk (how risky is this change, how confident is that estimate?), /summarize-behavior-changes (what will behave differently after this merges?), and /adversarial-pr-review (which likely failure can I reproduce?) dispatched in parallel with the sequential chain /analyze-test-coverage (is the change well covered by tests?), then /validate-pr (are we building the right product?), then /verify-pr (does it conform to the acceptance criteria?), then /review-pr (is the code well-crafted?).
When the diff touches a CLI, SDK, API, or database surface, the matching conditional surface review (/review-cli-changes, /review-sdk-changes, /review-api-changes, /review-db-changes) is dispatched alongside them (see Conditional surface reviews).
Each step posts its own report (updating that step's existing comment when there is one, see step 4) and marks the commit it reviewed.
When the run finishes, it commits this PR's report directory (reports plus any assets they reference, such as screenshots and casts) in the user-global reviews store and pushes it (see step 6).
/analyze-test-coverage runs as the first chain step so the coverage determination is explicit, tracked, and available before anything else judges the change: it is cheap static analysis, its report is posted before the rest of the chain starts, and the later steps (validate-pr, verify-pr, review-pr) can read it as evidence.
verify-pr and review-pr also embed parts of the coverage analysis in their own reports; the standalone step guarantees the introduced-tests / change-coverage / uncovered-code determination exists for the current commit even when the chain halts early.
It never halts the pipeline: a fail verdict (uncovered changes found) is reported in the summary for the human reviewer, who decides what to do with it.
Conditional surface reviews
Four surface reviews deep-dive one interface the diff touches: /review-cli-changes (commands, options, output, exit codes), /review-sdk-changes (public symbols, signatures, types, packaging), /review-api-changes (endpoints, fields, status codes, versioning), and /review-db-changes (migrations, locks, reversibility, backfills).
They are conditional: a surface review is stale only when the PR's changed files match that surface, so a PR that touches none of them dispatches none of them.
The staleness script decides applicability from the same file-path heuristics this skill documents (see step 4), so a run with a precomputed plan (--steps) and a run that computes its own plan select the same surface steps.
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.
- today Changed · +29 lines · +6 tokens per session 34e6a8eaf0de
- 9d ago Changed · +7 lines 484850e42534
- 11d ago Changed · +124 lines · +33 tokens per session 356df3ab98dd
- 14d ago Changed · +3 lines aecbe74d90e5
- 18d ago Changed · +30 lines · +7 tokens per session c6e6bd82d004
- 22d ago Changed · +65 lines e36b9861749e
- 24d ago Changed 6aab4b6cbe39
- 28d ago Changed · +9 lines 39f1347383e8
- 1mo ago Changed b37887730d67
- 1mo ago First seen · 360 lines · 62 tokens per session scan A a0d0ef0cb8b3
review-pr-full is a skill published in the GitHub repository tomzx/agents (13 stars, last pushed yesterday), licensed MIT. It adds 108 tokens to every session and 10,115 once invoked, about $0.0004 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 2026-09-03.
Other skills, from other repositories
pr
Draft a PR body from OMC's paper trail — the smallest visual that proves the change, evidence consumed from the verify protocol (never re-collected), reversibility from the ADR test on the diff, glossary language throughout.
merge-batch
Prepare and merge an owner-selected PR batch with contributor credit and exact-candidate admission. Use for a pasted PR list; combined focused validation is diagnostic, and every actual landing still requires its applicable checks and authorization.
otto-pr-review-loop
Run Polly and OCR on a PR until all findings settle.
engram-branch-pr
PR creation workflow for Engram following the issue-first enforcement system. Trigger: When creating a pull request, opening a PR, or preparing changes for review.
github-contributor
A step-by-step guide for contributing code, tests, documentation, or build changes to open-source projects you do not maintain. It explains how to prepare a pull request that follows the project’s rules and is easier for maintainers to review.
revdiff
Review diffs, files, and documents with inline annotations in a TUI overlay, or answer questions about revdiff usage, configuration, themes, and keybindings. Opens revdiff in agterm/tmux/zellij/herdr/kitty/wezterm/cmux/ghostty/iterm2/emacs-vterm, captures annotations, and addresses them. Works in git, hg, and jj repos…