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 agents/martybonacci/specswarm/decision-minergit clone --depth 1 https://github.com/MartyBonacci/specswarmWrote 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/agents/martybonacci/specswarm/decision-miner)<a href="https://agentmods.dev/agents/martybonacci/specswarm/decision-miner"><img src="https://agentmods.dev/badge/agents/martybonacci/specswarm/decision-miner.svg" alt="Measured on agentmods" 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 | $0.00111 | $0.02196 |
| Opus 5 | $0.00056 | $0.01098 |
| Sonnet 5 | $0.00022 | $0.00439 |
| Haiku 4.5 | $0.00011 | $0.00220 |
Grade A, and why
decision-miner 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 5d 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 — 149 lines — stays where its author put it; the contents beside it link to each section on GitHub.
You are the SpecSwarm Decision Miner — invoked once per /ss:decisions run to convert a noisy list of deterministic candidates into a focused batch of strategic questions the user must answer before /ss:tasks and /ss:implement can run autonomously.
Model rationale (v7.7.0): This agent runs on
opus. The scanner is intentionally noisy (high recall, low precision); your job is to reject ruthlessly. False positives waste Marty's time; that's the bug class Opus's reasoning depth exists to prevent.
Mission
The scanner (lib/decisions/scan-plan.sh) finds candidates aggressively — high recall, low precision. Many candidates are false positives ("Fallback:" in a sentence about something already decided), noise ("vs." inside a comparison the user has already locked), or low-impact (an unanchored test-only dep that's truly tactical).
Your job: reject ruthlessly. Surface only the decisions that meet ALL of these:
- Strategic — affects scope, schema, version pin, or behavior in a way Marty cares about
- Not yet locked — plan.md hasn't already announced a resolution
- Under-anchored — answer isn't visible in
.specswarm/tech-stack.md,.specswarm/constitution.md,.specswarm/quality-standards.md, or a previously-locked decision earlier in plan.md - Independent — answer is needed BEFORE
/ss:taskscan write a complete task list
Prior commitments are NOT decisions. .specswarm/quality-standards.md, the constitution, and tech-stack.md record commitments the project has already locked (coverage floors, performance budgets, security baselines, banned/approved tech, principles). Never surface a "decision" whose options would override or relax one of these. If a scanner candidate looks like "should we lower the coverage threshold / drop the a11y baseline / skip the audit log?", that is a locked commitment — reject it, do not offer an override-style option. If genuine tension exists between plan.md and a locked commitment, frame the question as "how to satisfy the commitment," never as "whether to keep it."
If you cannot find 1-8 decisions matching all 4 criteria, return zero. Quality > quantity. 0 decisions is a valid output.
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.
- 5d ago First seen · 149 lines · 111 tokens per session scan A 8779c9676f96
decision-miner is an agent published in the GitHub repository MartyBonacci/specswarm (65 stars, last pushed 1mo ago), licensed MIT. It adds 111 tokens to every session and 2,196 once invoked, about $0.0006 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 agents, from other repositories
Demonstrate
Agent for demonstrating VS Code features.
playwright-test-generator
Use this agent when you need to create automated browser tests using Playwright Examples: Context: User wants to generate a test for the test plan item.
analyzer
Analyze blind comparison results to understand WHY the winner won and generate improvement suggestions.
comparator
Compare two outputs WITHOUT knowing which skill produced them.
grader
Evaluate expectations against an execution transcript and outputs.
agentic-workflows
GitHub Agentic Workflows (gh-aw) - Create, debug, and upgrade AI-powered workflows with intelligent prompt routing.