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 marcoguillermaz/Tierward --skill pr-reviewgit clone --depth 1 https://github.com/marcoguillermaz/TierwardWrote 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/marcoguillermaz/tierward/pr-review)<a href="https://agentmods.dev/skills/marcoguillermaz/tierward/pr-review"><img src="https://agentmods.dev/badge/skills/marcoguillermaz/tierward/pr-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/marcoguillermaz/tierward/pr-review"><img src="https://agentmods.dev/badge/skills/marcoguillermaz/tierward/pr-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 | $0.00000 | $0.05759 |
| Opus 5.5 | $0.00000 | $0.02304 |
| Sonnet 5.5 | $0.00000 | $0.01152 |
| Haiku 4.5 | $0.00000 | $0.00576 |
Grade A, and why
pr-review scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 12d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 360 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Run an autonomous PR review locally. Classify findings by severity, post the review as a comment for audit trail, and surface a merge decision to the user. Read-only: the skill orchestrates a review; it does not modify code, does not push, does not merge.
Step 0 — Resolve repo + parse args
# Detect the GitHub repo from the local clone (no hard-coded org/repo).
REPO=$(gh repo view --json nameWithOwner -q .nameWithOwner)
If gh repo view fails: respond "This skill requires the gh CLI authenticated against a GitHub repo. Run \gh auth login` and ensure the project has a GitHub remote."` and stop.
Parse $ARGUMENTS:
| Arg | Behavior |
|---|---|
<PR_NUMBER> (positional, optional) |
Integer. If absent, detect from current branch in Step 1. |
--local |
Review the local working diff (git diff <base>...HEAD) instead of a PR. Use before the PR exists to catch findings pre-push. Auto-selected in Step 1 when no PR is found for the branch, so an explicit flag is only needed to force local mode while an open PR also exists. |
--base <branch> |
Base branch for --local diff (default: the repo's default branch, resolved in Step 1). Ignored in PR mode (the PR carries its own base). |
--deep |
Escalate the review subagent from sonnet to opus. Use for changes touching auth, money, migrations, or shared utilities. Adds ~30-60s latency. |
--with-context |
Pass the active session file (.claude/session/block-*.md if present) to the review subagent. Default OFF — review stays diff-pure and unbiased by prior decisions. |
Examples:
/pr-review 122/pr-review 122 --deep/pr-review --deep(resolves PR from current branch; falls back to local diff if none open)/pr-review --local(review the working diff against the default base, before opening a PR)/pr-review --local --base staging
Step 1 — Resolve PR number, or select local-diff mode
Mode selection:
- If
--localwas passed → local mode (skip PR resolution entirely). - Else if
PR_NUMBERwas passed → PR mode with that number. - Else detect a PR for the current branch:
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 Changed · +1 lines 403f6aeaeab4
- 15d ago Changed aedbfa841c73
- 16d ago Changed · +3 lines 7b7c43fa09ae
- 1mo ago First seen · 356 lines · 0 tokens per session scan A b3a48b0f6a78
pr-review is a skill published in the GitHub repository marcoguillermaz/Tierward (4 stars, last pushed 2d ago), licensed MIT. It costs nothing until one of its globs matches a file; then it loads 5,759 tokens. 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
github-pr-review
Review one exact GitHub pull-request head from bounded changed-file patches and check evidence, producing source-line findings without posting or claiming tests ran.
ensemble-beads-build
Drive an existing bead hierarchy to completion through the full builder, code-review, and close pipeline (Codex skill for /ensemble:beads-build).
ensemble-discover-standards
Analyze codebase and extract coding conventions into a standards/ directory with index.yml (Codex skill for /ensemble:discover-standards).
revidere
A Git branch review workflow that groups changes into sections, labels their importance and whether they need attention, and checks that every changed line is explained. It creates a review page with selected differences and links to GitHub.
address-conductor-comment
A workflow for handling pending Conductor code-review comments, including reading their discussion and assigning changes to subagents when appropriate.
explain-comment
A skill that reads a specified file or code range and adds explanatory comments in Japanese. The comments use a question style and focus on important logic, design choices, and side effects.