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 OutlineDriven/odin-claude-plugin --skill make-pr-easy-to-reviewgit clone --depth 1 https://github.com/OutlineDriven/odin-claude-pluginWrote 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/outlinedriven/odin-claude-plugin/make-pr-easy-to-review)<a href="https://agentmods.dev/skills/outlinedriven/odin-claude-plugin/make-pr-easy-to-review"><img src="https://agentmods.dev/badge/skills/outlinedriven/odin-claude-plugin/make-pr-easy-to-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/outlinedriven/odin-claude-plugin/make-pr-easy-to-review"><img src="https://agentmods.dev/badge/skills/outlinedriven/odin-claude-plugin/make-pr-easy-to-review.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 29 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 | $0.00036 | $0.01197 |
| Opus 5 | $0.00018 | $0.00598 |
| Sonnet 5 | $0.00007 | $0.00239 |
| Haiku 4.5 | $0.00004 | $0.00120 |
Grade A, and why
make-pr-easy-to-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 5d 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 — 46 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Make PR easy to review
Contract
| Field | Bound contract |
|---|---|
| Trigger | A human explicitly invokes this skill to reshape or annotate one pull request for review. |
| Authority | Human-gated: previews the exact commits and consequences and receives explicit human consent before rewriting the named local branch history; otherwise reversible local: inspects and annotates the named pull request locally and writes only its local branch; rollback is version control. Never pushes, publishes, alters credentials, or mutates any remote. |
| Side effect | May rewrite only the named local branch history. Preserve a recovery ref and use version control to restore the original history if recovery is needed. |
| Done | The review surface accurately emphasizes risk, the revised commit sequence is coherent, and the final tree identity exactly matches the pre-rewrite intended tree. |
Inputs
Required: the pull request or local branch, its base ref, and the human's requested review outcome. The branch and base must resolve locally, and the working tree must have no uncommitted changes that could be confused with the pull request. If history rewriting is proposed, explicit consent is required after the rewrite preview; annotation-only work does not require rewrite consent.
Procedure
- Resolve the named branch and base ref. Record the branch tip, merge-base, commit range, changed paths, and final tree identifier. Stop if the target is ambiguous, the base is unavailable, or unrelated working-tree changes are present. Done when: the branch tip, merge-base, commit range, changed paths, and tree identifier are all recorded.
- Read the complete diff and commit sequence. Identify behavioral changes, data or migration effects, security and trust boundaries, concurrency or lifecycle changes, compatibility breaks, and tests or evidence tied to those risks. Do not invent risk or verification evidence. Done when: every risk category is assessed and each identified risk is mapped to its paths and evidence.
- Draft a review surface that states the pull request's purpose, orders the risky areas first, maps each risk to the relevant paths and available evidence, and calls out unresolved or unverified claims explicitly. Done when: the review surface is drafted with risks ordered first and every claim mapped to evidence or marked unverified.
- Decide whether annotation alone provides an accurate review surface. If so, return the proposed title, description, review order, and risk notes without changing history. Done when: the annotation-only decision is made and, if yes, the result is returned.
- If reshaping history would materially improve reviewability, design a commit sequence in which each commit has one coherent purpose, preserves dependency order, and contains the tests or evidence that belong to its change. Preview the exact source commit range, proposed sequence, local branch to be rewritten, recovery ref, and the fact that commit identifiers will change. Done when: the proposed sequence and its preview are complete.
- Obtain explicit human consent to that exact rewrite. Without consent, make no history change and return the annotation-only result plus the declined rewrite preview. Done when: consent is obtained or the annotation-only result is returned with the declined preview.
- Create a recovery ref at the original branch tip, then perform only the approved local rewrite. Do not widen the changed-path set or alter the intended final content while splitting, reordering, or combining commits. Done when: the rewrite is performed and the recovery ref exists.
- Recompute the rewritten branch's final tree identifier and compare it byte-for-byte with the identifier recorded in step 1. Also inspect the rewritten commit range to confirm that its order and messages match the approved sequence. Done when: the tree identifiers match and the commit sequence matches the approved preview.
- If tree identity differs, stop, restore the original branch tip from the recovery ref, and report the mismatch. If it matches, retain the recovery ref, report the old and new tip identifiers and the identical tree identifier, and return the risk-oriented review surface. Do not push the rewritten branch. Done when: the result is returned with the tree-identity verdict and the review surface.
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.
- 5d ago First seen · 46 lines · 36 tokens per session scan A 5fe4231e0886
make-pr-easy-to-review is a skill published in the GitHub repository OutlineDriven/odin-claude-plugin (36 stars, last pushed 3d ago), licensed Apache-2.0. It adds 36 tokens to every session and 1,197 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 2026-09-06.
Other skills, from other repositories
slop-scan
Use when scanning tracked repository files for AI slop, verbosity, brittle references, or low-value contributions.
weave-fix-review
Fix weave review findings — validate, add test coverage, fix, and commit each as atomic changes.
action
Use when screened review findings should be fixed on the current branch, usually as one verified commit per finding.
goal
Use when the review-and-CI loop should continue through bounded rounds until no new actionable feedback remains.
systematic-debugging
Use when encountering any bug, test failure, or unexpected behavior, before proposing fixes.
local-ai-agents
Build local-first AI agents that run entirely on a developer workstation with Microsoft Foundry Local and Qwen function-calling models. Covers Small Language Models (SLMs), the OpenAI-compatible local endpoint, sandboxed local tools, local RAG with Chroma, local MCP servers, hybrid cloud/local routing, and the…