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.
git clone --depth 1 https://github.com/cloudnative-co/claude-code-starter-kitWrote 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/commands/cloudnative-co/claude-code-starter-kit/update-kit-dry-run)<a href="https://agentmods.dev/commands/cloudnative-co/claude-code-starter-kit/update-kit-dry-run"><img src="https://agentmods.dev/badge/commands/cloudnative-co/claude-code-starter-kit/update-kit-dry-run/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/commands/cloudnative-co/claude-code-starter-kit/update-kit-dry-run"><img src="https://agentmods.dev/badge/commands/cloudnative-co/claude-code-starter-kit/update-kit-dry-run.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 | $0.02179 |
| Opus 5.5 | $0.00000 | $0.00872 |
| Sonnet 5.5 | $0.00000 | $0.00436 |
| Haiku 4.5 | $0.00000 | $0.00218 |
Grade A, and why
update-kit-dry-run scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 10d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 190 lines — stays where its author put it; the contents beside it link to each section on GitHub.
/update-kit-dry-run - Preview Update Changes
Preview what /update-kit would change without actually deploying.
Instructions
Run the following block as one Bash invocation. It uses the same strict,
non-evaluating manifest/config lookup as /update-kit: a current non-MDM
manifest binds the checkout to the exact config used at install time, while a
legacy or MDM manifest uses ~/.claude-starter-kit.conf. A legacy config with
no KIT_REPO key falls back to ~/.claude-starter-kit. An invalid explicit
binding, key, or repository stops the preview instead of silently using another
path.
set -euo pipefail
default_config_file="$HOME/.claude-starter-kit.conf"
config_file="$default_config_file"
default_kit_repo="$HOME/.claude-starter-kit"
kit_repo="$default_kit_repo"
manifest_file="$HOME/.claude/.starter-kit-manifest.json"
manifest_bound=false
if [ -e "$manifest_file" ] || [ -L "$manifest_file" ]; then
if [ -L "$manifest_file" ] || [ ! -f "$manifest_file" ]; then
printf '%s\n' "Invalid starter-kit manifest: $manifest_file" >&2
exit 1
fi
manifest_binding="$(jq -cse '
if length == 1 and (.[0] | type == "object") then
.[0]
| if (has("kit_repo") or has("config_file")) then
if (has("kit_repo") and has("config_file")
and (.mdm_managed == false)
and ((.kit_repo | type) == "string")
and ((.config_file | type) == "string")
and (.kit_repo | startswith("/"))
and (.config_file | startswith("/"))
and ((.kit_repo | test("[\\x00-\\x1f\\x7f]")) | not)
and ((.config_file | test("[\\x00-\\x1f\\x7f]")) | not))
then [.kit_repo, .config_file]
else error("invalid runtime binding")
end
else []
end
else error("invalid manifest")
end
' "$manifest_file" 2>/dev/null)" || {
printf '%s\n' "Invalid starter-kit manifest binding" >&2
exit 1
}
if [ "$manifest_binding" != "[]" ]; then
kit_repo="$(printf '%s' "$manifest_binding" | jq -r '.[0]')" || exit 1
config_file="$(printf '%s' "$manifest_binding" | jq -r '.[1]')" || exit 1
manifest_bound=true
fi
fi
if [ -e "$config_file" ] || [ -L "$config_file" ]; then
if [ -L "$config_file" ] || [ ! -f "$config_file" ] \
|| [ ! -r "$config_file" ]; then
printf '%s\n' "Invalid starter-kit config: $config_file" >&2
exit 1
fi
kit_repo_count="$(awk '
/^[[:space:]]*KIT_REPO[[:space:]]*=/ { count++ }
END { print count + 0 }
' "$config_file")" || {
printf '%s\n' "Could not read starter-kit config" >&2
exit 1
}
if [ "$kit_repo_count" -gt 1 ]; then
printf '%s\n' "Multiple KIT_REPO entries in starter-kit config" >&2
exit 1
fi
if [ "$kit_repo_count" -eq 1 ]; then
kit_repo_line="$(awk '
/^[[:space:]]*KIT_REPO[[:space:]]*=/ { print }
' "$config_file")" || exit 1
if [ "${kit_repo_line#KIT_REPO=\"}" = "$kit_repo_line" ] \
|| [ "${kit_repo_line%\"}" = "$kit_repo_line" ]; then
printf '%s\n' "Invalid KIT_REPO entry in starter-kit config" >&2
exit 1
fi
config_kit_repo="${kit_repo_line#KIT_REPO=\"}"
config_kit_repo="${config_kit_repo%\"}"
case "$config_kit_repo" in
/*) ;;
*)
printf '%s\n' "Invalid KIT_REPO entry in starter-kit config" >&2
exit 1
;;
esac
case "$config_kit_repo" in
*'"'*)
printf '%s\n' "Invalid KIT_REPO entry in starter-kit config" >&2
exit 1
;;
esac
if printf '%s' "$config_kit_repo" | LC_ALL=C grep -q '[[:cntrl:]]'; then
printf '%s\n' "Invalid KIT_REPO entry in starter-kit config" >&2
exit 1
fi
if [ "$manifest_bound" = true ] && [ "$config_kit_repo" != "$kit_repo" ]; then
printf '%s\n' "Manifest binding does not match starter-kit config" >&2
exit 1
fi
kit_repo="$config_kit_repo"
elif [ "$manifest_bound" = true ]; then
printf '%s\n' "Manifest-bound config has no KIT_REPO entry" >&2
exit 1
fi
elif [ "$manifest_bound" = true ]; then
printf '%s\n' "Manifest-bound config not found: $config_file" >&2
exit 1
fi
kit_repo_physical="$(cd "$kit_repo" 2>/dev/null && pwd -P)" || {
printf '%s\n' "Starter-kit checkout not found: $kit_repo" >&2
exit 1
}
repo_top="$(git -C "$kit_repo_physical" rev-parse --show-toplevel 2>/dev/null)" || {
printf '%s\n' "Not a Git checkout: $kit_repo_physical" >&2
exit 1
}
repo_top_physical="$(cd "$repo_top" 2>/dev/null && pwd -P)" || exit 1
if [ "$kit_repo_physical" != "$repo_top_physical" ] \
|| [ ! -e "$kit_repo_physical/.git" ] \
|| [ ! -f "$kit_repo_physical/setup.sh" ] \
|| [ ! -f "$kit_repo_physical/lib/features.sh" ] \
|| [ ! -f "$kit_repo_physical/config/plugins.json" ]; then
printf '%s\n' "KIT_REPO is not a starter-kit repository root" >&2
exit 1
fi
printf 'Resolved kit repo: %s\n' "$kit_repo_physical"
printf 'Resolved config file: %s\n' "$config_file"
repo_status="$(git -C "$kit_repo_physical" status --porcelain)" || {
printf '%s\n' "Could not inspect starter-kit checkout" >&2
exit 1
}
if [ -n "$repo_status" ]; then
printf '%s\n' "Local changes found; review them before previewing" >&2
exit 1
fi
git -C "$kit_repo_physical" fetch --tags
git -C "$kit_repo_physical" pull --ff-only
setup_args=(--update --dry-run)
if [ "$manifest_bound" = true ]; then
setup_args+=("--config=$config_file")
fi
(cd "$kit_repo_physical" && bash setup.sh "${setup_args[@]}")
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.
- 10d ago Changed · +1 lines 506e885d9104
- 1mo ago First seen · 189 lines · 0 tokens per session scan A 5fcee89c4f67
update-kit-dry-run is a command published in the GitHub repository cloudnative-co/claude-code-starter-kit (153 stars, last pushed 6d ago), licensed MIT. It costs nothing until one of its globs matches a file; then it loads 2,179 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 2026-08-30.
Other commands, from other repositories
checklist
Generate a custom checklist for the current feature based on user requirements.
clarify
Identify underspecified areas in the current feature spec by asking up to 5 highly targeted clarification questions and encoding answers back into the spec.
specify
Create or update the feature specification from a natural language feature description.
converge
Assess the current codebase against the feature's spec, plan, and tasks, then append any remaining unbuilt work as new tasks to tasks.md so implement can complete it.
implement
Execute the implementation plan by processing and executing all tasks defined in tasks.md.
speckit.bug.fix
Apply the remediation from a bug assessment and record what was changed.