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/martybonacci/specswarm/decisionsgit clone --depth 1 https://github.com/MartyBonacci/specswarmWhat 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 | $0.00074 | $0.03865 |
| Opus 5 | $0.00037 | $0.01932 |
| Sonnet 5 | $0.00015 | $0.00773 |
| Haiku 4.5 | $0.00007 | $0.00386 |
Grade A, and why
decisions scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 3d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 343 lines — stays where its author put it; the contents beside it link to each section on GitHub.
SpecSwarm Decision Pre-Batching
The biggest Marty-time reducer in the v7.x toolchain. Mid-chunk strategic-decision interrupts cost ~15–25 interactions across /ss:tasks + /ss:implement. This command collapses them into ONE front-loaded touchpoint by:
- Deterministically scanning
plan.mdfor candidate decisions - Dispatching the
decision-minersubagent to dedup/prioritize/polish into 0–8 well-formed questions - Asking Marty all of them via 1-2
AskUserQuestioncalls (4-question limit per call) - Writing
decision-sheet.md(canonical record + provenance) - Appending a "Pre-Batched Decisions" section to
plan.mdso/ss:tasksand/ss:implementsee the answers without code changes
After this command runs successfully, the chunk should execute autonomously through /ss:tasks → /ss:implement with minimal further interrupts (only true edge cases the miner missed).
Phase 1: Resolve feature
PLUGIN_DIR="$(cd "$(dirname "${BASH_SOURCE[0]}")/.." && pwd)"
# shellcheck disable=SC1091
source "${PLUGIN_DIR}/lib/features-location.sh"
# shellcheck disable=SC1091
source "${PLUGIN_DIR}/lib/decisions/scan-plan.sh"
# shellcheck disable=SC1091
[ -f "${PLUGIN_DIR}/lib/notify.sh" ] && source "${PLUGIN_DIR}/lib/notify.sh"
FEATURE_NUM=""
SCAN_ONLY=false
DRY_RUN=false
while [ $# -gt 0 ]; do
case "$1" in
--scan-only) SCAN_ONLY=true; shift ;;
--dry-run) DRY_RUN=true; shift ;;
-h|--help)
cat <<EOF
Usage: /ss:decisions [FEATURE_NUM] [options]
Pre-batch strategic decisions for a feature.
Options:
--scan-only Print deterministic candidates only (no agent, no questions)
--dry-run Scan + agent (writes decision-sheet.draft.md); skip questions + plan.md mutation
Examples:
/ss:decisions
/ss:decisions 002
/ss:decisions --scan-only
/ss:decisions --dry-run
EOF
exit 0
;;
*)
if [ -z "$FEATURE_NUM" ]; then FEATURE_NUM="$1"; fi
shift
;;
esac
done
REPO_ROOT=$(git rev-parse --show-toplevel 2>/dev/null || pwd)
# Resolve feature (same cascade as /ss:retrospective)
if [ -z "$FEATURE_NUM" ]; then
BRANCH=$(git rev-parse --abbrev-ref HEAD 2>/dev/null || echo "")
FEATURE_NUM=$(echo "$BRANCH" | grep -oE '^[0-9]{3}' || echo "")
fi
if [ -z "$FEATURE_NUM" ]; then
get_features_dir "$REPO_ROOT"
FEATURE_NUM=$(find "$FEATURES_DIR" -maxdepth 1 -type d -name '[0-9][0-9][0-9]-*' 2>/dev/null \
| sort | tail -1 | xargs -n1 basename 2>/dev/null \
| grep -oE '^[0-9]{3}' || echo "")
fi
if [ -z "$FEATURE_NUM" ]; then
echo "❌ No feature number provided and no feature dirs found." >&2
exit 2
fi
if ! find_feature_dir "$FEATURE_NUM" "$REPO_ROOT"; then
echo "❌ Feature $FEATURE_NUM not found." >&2
exit 2
fi
FEATURE_ID=$(basename "$FEATURE_DIR")
PLAN_PATH="${FEATURE_DIR}/plan.md"
if [ ! -f "$PLAN_PATH" ]; then
echo "❌ plan.md not found at $PLAN_PATH. Run /ss:plan first." >&2
exit 2
fi
FOUNDATION="${REPO_ROOT}/.specswarm"
TECH_STACK="${FOUNDATION}/tech-stack.md"
CONSTITUTION="${FOUNDATION}/constitution.md"
QUALITY_STANDARDS="${FOUNDATION}/quality-standards.md"
echo "🔍 Decisions for $FEATURE_ID"
echo " plan.md: $PLAN_PATH"
echo " tech-stack.md: $([ -f "$TECH_STACK" ] && echo "$TECH_STACK" || echo "(absent — version-anchor check skipped)")"
echo " constitution: $([ -f "$CONSTITUTION" ] && echo "$CONSTITUTION" || echo "(absent — constitution callout check skipped)")"
echo " quality-standards:$([ -f "$QUALITY_STANDARDS" ] && echo " $QUALITY_STANDARDS" || echo " (absent — prior-commitments check skipped)")"
echo ""
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.
- 3d ago First seen · 343 lines · 74 tokens per session scan A efbe2d974ba8
decisions is a command published in the GitHub repository MartyBonacci/specswarm (65 stars, last pushed 1mo ago), licensed MIT. It adds 74 tokens to every session and 3,865 once invoked, about $0.0004 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 commands, from other repositories
git
Git operations with intelligent commit messages and workflow optimization.
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.
analyze
Perform a non-destructive cross-artifact consistency and quality analysis across spec.md, plan.md, and tasks.md after task generation.
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.