Getting it into your agent
This one installs as part of its plugin. Adding the marketplace and installing the plugin brings it with everything else the plugin ships.
/plugin marketplace add gwittebolle/claude-carbon/plugin install claude-carbonWrote 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/gwittebolle/claude-carbon/carbon-pr)<a href="https://agentmods.dev/skills/gwittebolle/claude-carbon/carbon-pr"><img src="https://agentmods.dev/badge/skills/gwittebolle/claude-carbon/carbon-pr/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/gwittebolle/claude-carbon/carbon-pr"><img src="https://agentmods.dev/badge/skills/gwittebolle/claude-carbon/carbon-pr.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 | $0.00016 | $0.00862 |
| Opus 5 | $0.00008 | $0.00431 |
| Sonnet 5 | $0.00003 | $0.00172 |
| Haiku 4.5 | $0.00002 | $0.00086 |
Grade A, and why
carbon-pr scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 8d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
What it actually says
Run the following bash script from the repository the PR belongs to, and present the output to the user.
#!/usr/bin/env bash
# Locate the install wired into the status line (honours CLAUDE_CONFIG_DIR),
# then fall back to CLAUDE_PLUGIN_ROOT / CLAUDE_CARBON_DIR / the default path.
# Windows (Git Bash): CLAUDE_* variables and the paths stored in settings.json can
# arrive in the native "C:\Users\..." spelling, which bash cannot open. cygpath
# ships with Git for Windows; everywhere else this is the identity function.
ccp() {
case "${OSTYPE:-}" in msys*|cygwin*) ;; *) printf '%s' "$1"; return 0 ;; esac
case "$1" in
[A-Za-z]:[\\/]*|*\\*) cygpath -u "$1" 2>/dev/null || printf '%s' "$1" ;;
*) printf '%s' "$1" ;;
esac
}
CFG="$(ccp "${CLAUDE_CONFIG_DIR:-$HOME/.claude}")"
REPO_DIR=""
if command -v jq >/dev/null 2>&1 && [ -f "$CFG/settings.json" ]; then
SL_CMD="$(jq -r '.statusLine.command // empty' "$CFG/settings.json" 2>/dev/null)"
# statusLine.command stores a shell-escaped path: expand ~ and unescape spaces
SL_CMD="${SL_CMD//\\ / }"; SL_CMD="${SL_CMD/#\~/$HOME}"; SL_CMD="$(ccp "$SL_CMD")"
[ -n "$SL_CMD" ] && [ -f "$SL_CMD" ] && REPO_DIR="$(cd "$(dirname "$SL_CMD")/.." 2>/dev/null && pwd)"
fi
[ -z "$REPO_DIR" ] && [ -n "${CLAUDE_PLUGIN_ROOT:-}" ] && REPO_DIR="$(ccp "$CLAUDE_PLUGIN_ROOT")"
[ -z "$REPO_DIR" ] && [ -n "${CLAUDE_CARBON_DIR:-}" ] && REPO_DIR="$(ccp "$CLAUDE_CARBON_DIR")"
if [ -z "$REPO_DIR" ]; then
# Pure marketplace installs reach here: no statusLine.command in settings.json and
# no CLAUDE_PLUGIN_ROOT in a skill's bash env. Pick the newest cached plugin copy,
# keeping only candidates that ship the PR report script.
CACHE_LATEST=""
for D in "$CFG/plugins/cache"/*/claude-carbon/*/; do
[ -f "${D}scripts/generate-pr-report.sh" ] || continue
if [ -z "$CACHE_LATEST" ] || [ "$(printf '%s\n%s\n' "$CACHE_LATEST" "$D" | sort -V | tail -1)" = "$D" ]; then
CACHE_LATEST="$D"
fi
done
[ -n "$CACHE_LATEST" ] && REPO_DIR="${CACHE_LATEST%/}"
fi
[ -z "$REPO_DIR" ] && REPO_DIR="$HOME/code/claude-carbon"
bash "$REPO_DIR/scripts/generate-pr-report.sh"
The script sums the local Claude Code sessions recorded for the current branch of the current project and posts (or updates in place) one sticky comment on the branch's open PR, through the user's own gh auth. Report its final line to the user as printed; never recompute or reformat the numbers.
If it reports no sessions for the branch, tell the user that only sessions recorded since the git_branch column exists carry a branch, and that scripts/backfill.sh repairs older rows while their transcripts are still on disk. If it reports no open PR, suggest pushing the branch and opening one first, or passing --pr <number>. If the script itself is missing (an install that predates it), suggest /carbon-update.
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.
- 8d ago Changed · +10 lines e02ae6c3c306
- 11d ago First seen · 43 lines · 16 tokens per session scan A 06f0e4231e89
carbon-pr is a skill published in the GitHub repository gwittebolle/claude-carbon (193 stars, last pushed yesterday), licensed MIT. It adds 16 tokens to every session and 862 once invoked, about $0.0001 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 2026-08-30.
Other skills, from other repositories
find-simplifications
Use for a periodic repo-wide sweep of qwen-code for accumulated excess surface — dead components and files, orphaned locale keys, exports nothing consumes, added-then-removed scaffolding — filing candidates on a tracking issue and landing only what a maintainer has said yes to. Repo-wide and evidence-first; every…
swarm-pr-review
Run a graph-guided, tool-augmented PR review using context packing, parallel exploration, mandatory repository-agnostic risk-family coverage with dispatch scaled to diff size and risk, independent reviewer validation, critic challenge, and metrics writeback. Use for deep pull request review with low false-positive…
review
Validate plans, execution, or PRs against wish criteria — returns SHIP / FIX-FIRST / BLOCKED with severity-tagged gaps.
include-test-files-that-assert-on-behavior-being-changed-in-decl
When delegating a task affected by this skill, include.
review
Adversarial fresh-context review of an increment before it ships. Every finding cites path:line and is re-verified. Use when saying "review", "grill this", or "critique the implementation".
tech-debt-ci-review
Codex adapter for deep technical-debt and CI-stability audits. Use when asked to find test theater, flaky tests, missing or mis-scoped tests, brittle CI/toolchain behavior, structural debt blocking green PRs, or a remediation order for opencode-swarm.