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 agentmods add commands/callmetechie/fleet-manager/compose-upgit clone --depth 1 https://github.com/CallMeTechie/fleet-managerWhat 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 | $0.00038 | $0.00844 |
| Opus 5 | $0.00019 | $0.00422 |
| Sonnet 5 | $0.00008 | $0.00169 |
| Haiku 4.5 | $0.00004 | $0.00084 |
Grade A, and why
compose-up scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 2d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
What it actually says
Compose Up
set -euo pipefail
source "${CLAUDE_PLUGIN_ROOT:-plugin}/commands/_compose-lib.sh"
FILE=""; PROJECT=""; SERVER_ARG=""
set -f # no globbing of unquoted $ARGUMENTS (review H2)
# shellcheck disable=SC2086 # intentional: split ARGUMENTS into positional params
set -- ${ARGUMENTS:-}
while [ $# -gt 0 ]; do
case "$1" in
--file) shift; FILE="${1:-}" ;;
--file=*) FILE="${1#*=}" ;;
--*) echo "Unknown flag: $1" >&2; exit 1 ;;
*) if [ -z "$SERVER_ARG" ]; then SERVER_ARG="$1"; else PROJECT="$1"; fi ;;
esac
shift
done
# Single positional = project (active server assumed)
if [ -n "$SERVER_ARG" ] && [ -z "$PROJECT" ] && [ -z "$FILE" ]; then PROJECT="$SERVER_ARG"; SERVER_ARG=""; fi
# Exactly one of PROJECT / FILE
if { [ -n "$PROJECT" ] && [ -n "$FILE" ]; } || { [ -z "$PROJECT" ] && [ -z "$FILE" ]; }; then
echo "Usage: /compose-up [server] <project> OR /compose-up [server] --file /abs/path.yml" >&2
exit 1
fi
NAME="$(resolve_server "$SERVER_ARG")"
build_ssh "$NAME"; echo_target "$NAME"
is_scope_authorized "$NAME" docker_compose || {
echo "REFUSED: scope 'docker_compose' not authorized for '$NAME'. Tick it in the profile (or re-run /add-server)." >&2; exit 1; }
resolve_docker_cmd "$NAME"; docker_precheck
if [ -n "$FILE" ]; then
if ! [[ "$FILE" =~ ^/[A-Za-z0-9/_.-]+\.ya?ml$ ]]; then
echo "ERROR: --file needs an ABSOLUTE path to a .yml/.yaml file (got '$FILE')." >&2; exit 1
fi
case "$FILE" in *..*) echo "ERROR: --file path must not contain '..' (no traversal)." >&2; exit 1 ;; esac
if "${FM_SSH[@]}" "test -f '$FILE'"; then :; else
rc=$?
if [ "$rc" -eq 255 ]; then echo "ERROR: SSH to $NAME failed." >&2; else echo "ERROR: '$FILE' not found on $NAME." >&2; fi
exit 1
fi
"${FM_SSH[@]}" "$FM_DOCKER compose -f $FILE up -d"
else
ENTRY="$(compose_find_project "$PROJECT")"
if [ -z "$ENTRY" ]; then
echo "ERROR: project '$PROJECT' is not indexed yet (never started)." >&2
echo " For a first deploy use: /compose-up $NAME --file /abs/path/docker-compose.yml" >&2
exit 1
fi
CFG_ARGS="$(compose_config_args "$(echo "$ENTRY" | jq -r '.ConfigFiles')")"
# shellcheck disable=SC2086
"${FM_SSH[@]}" "$FM_DOCKER compose $CFG_ARGS up -d"
fi
echo ""; echo "Verdict:"; compose_ls_json | jq -r '.[] | [.Name, .Status] | @tsv' || echo "(state re-query failed — action was sent)"
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.
- 2d ago First seen · 65 lines · 38 tokens per session scan A 7497c3e5e6e0
compose-up is a command published in the GitHub repository CallMeTechie/fleet-manager (1 stars, last pushed 29d ago), licensed MIT. It adds 38 tokens to every session and 844 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 2026-08-31.
Other commands, from other repositories
log-fix
Read Docker/installer logs and iteratively fix bugs based on log analysis.
add-new-service
Add a new optional service called $ARGUMENTS to the Selfhost AI project.
containers
Apply the CLI runtime and safety contract before executing a command. Runtime --help is authoritative.
deploy-local-image
Apply the CLI runtime and safety contract before executing a command. Runtime --help is authoritative.
deploy-image
Apply the CLI runtime and safety contract before executing a command. Runtime --help is authoritative.
deploy
You are tasked with deploying the mcp-server-langgraph application to various targets. This command.