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 magic5644/Graph-It-Live --skill pr-reviewgit clone --depth 1 https://github.com/magic5644/Graph-It-LiveWrote 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/magic5644/graph-it-live/pr-review)<a href="https://agentmods.dev/skills/magic5644/graph-it-live/pr-review"><img src="https://agentmods.dev/badge/skills/magic5644/graph-it-live/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/magic5644/graph-it-live/pr-review"><img src="https://agentmods.dev/badge/skills/magic5644/graph-it-live/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 | $0.00132 | $0.02187 |
| Opus 5.5 | $0.00053 | $0.00875 |
| Sonnet 5.5 | $0.00026 | $0.00437 |
| Haiku 4.5 | $0.00013 | $0.00219 |
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 6d 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 — 195 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Graph-It-Live PR Review
Use deterministic local diff analysis first. Deepen only findings that need additional evidence.
CLI and Capability Discovery
Prefer the Graph-It CLI. At the first activation of any Graph-It skill in a session, check
graph-it --version when the CLI and terminal are available; use only its existing newer-version
notice and do not poll npm. Share discovery, version, and update decisions across Graph-It skills
for the session. Run graph-it tool --list once, consult graph-it --help or command help as
needed, and refresh only when the CLI/version, server, or workspace changes or a requested tool is
unknown. Examples are illustrative, not a fixed inventory. Do not install automatically. If the
CLI is unavailable or unsuitable, use MCP only when the connected host advertises a structurally
equivalent tool and its schema; otherwise explain the gap. graph-it tool <name> invokes analysis
directly through CLI; graph-it serve starts MCP. CLI --format and MCP response_format are
distinct.
If the CLI reports a newer version, interactive use requires explicit positive confirmation before
running graph-it update; silence, timeout, refusal, non-TTY, or generic Agent mode is not consent.
When automatic mode is explicitly enabled and pre-authorized, default the update decision to yes and
run graph-it update without another prompt. After success, refresh the version and --list,
invalidate cached capabilities, and advise that a running MCP server needs a host-authorized restart.
On failure, keep using the existing CLI only if it remains usable; otherwise explain the failure
and stop CLI analysis. Warn briefly and do not retry in a loop. Do not install the CLI without an
explicit request.
Fetch the Git base ref needed for review. The CLI indexes automatically for review-pr, so do not
run graph-it scan first.
Review from the repository (monorepo) root so consumers in every package are counted: pass it
explicitly with graph-it -w <repoRoot> review-pr ... rather than relying on cd. An explicit
-w is used exactly as given; without it the CLI uses the nearest directory holding package.json
or tsconfig.json, which inside a sub-package leaves sibling packages out. The index and cache
(.graph-it/) are created in that root. If stderr warns that imports resolve outside the workspace
root, impact on the packages outside it is not covered: rerun with the --workspace it names or
state the limitation in the review.
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.
- 6d ago Changed · +8 lines c604cf22cab2
- 13d ago Changed · +37 lines 00fecfd71e8a
- 14d ago Changed fee494cf8669
- 18d ago Changed · +1 lines 7ee84a57e504
- 1mo ago First seen · 149 lines · 132 tokens per session scan A 35c156a3a944
pr-review is a skill published in the GitHub repository magic5644/Graph-It-Live (61 stars, last pushed today), licensed MIT. It adds 132 tokens to every session and 2,187 once invoked, about $0.0005 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 2026-09-13.
Other skills, from other repositories
git-pr-workflows-git-workflow
Guide changes through code checks, testing, commits, and pull request preparation. Use when coordinating a complete Git workflow and verifying readiness for collaboration.
pr-review
Address feedback left on a GitHub pull request: fetch unresolved review threads, make agreed Elixir/Phoenix code fixes, reply, and resolve. Use for a PR URL/number or reviewer comments. NOT for pre-PR review, findings triage, or CI monitoring.
phx-pr-review
Address feedback left on a GitHub pull request: fetch unresolved review threads, make agreed Elixir/Phoenix code fixes, reply, and resolve. Use for a PR URL/number or reviewer comments. NOT for pre-PR review, findings triage, or CI monitoring.
signing-preflight
Before the first commit of a session, check that commit signing will work — the key is loaded and unlocked — and ask the operator to unlock it once, up front, instead of discovering it after a hundred unsigned commits. Never disables signing silently, never re-signs history without asking. Use at the start of…
loop-gh-pr
A recipe for a scheduled GitHub pull-request watcher. It checks open pull requests for new activity and prepares read-only review drafts without posting comments or merging anything.
review
Structured code review with parallel audit agents, confidence-scored triage, and optional auto-fix. Examines uncommitted changes, staged diffs, commit ranges, or specific paths. Produces a tiered report (MUST-FIX / RECOMMENDED / NIT) backed by evidence, then optionally applies fixes with verification.