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/marcusgoll/spec-flow/clarifygit clone --depth 1 https://github.com/marcusgoll/Spec-FlowWhat 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 | $0.00024 | $0.06021 |
| Opus 5 | $0.00012 | $0.03010 |
| Sonnet 5 | $0.00005 | $0.01204 |
| Haiku 4.5 | $0.00002 | $0.00602 |
Grade A, and why
clarify 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.
How it starts
The opening of the file, as written. The whole thing — 801 lines — stays where its author put it; the contents beside it link to each section on GitHub.
/clarify — Specification Clarifier (Thin Wrapper)
v11.0 Architecture: This command spawns the isolated
clarify-phase-agentvia Task(). All clarification logic runs in isolated context with question batching.
Active Feature: !ls -td specs/[0-9]*-* 2>/dev/null | head -1 || echo "none"
Interaction State: !cat specs/*/interaction-state.yaml 2>/dev/null | head -10 || echo "none"
Architecture (v11.0 - Phase Isolation):
/clarify → Task(clarify-phase-agent) → Q&A loop if needed → updated spec.md
Agent responsibilities:
- Scan spec.md for
[NEEDS CLARIFICATION]markers - Map ambiguities to question bank templates
- Return batched questions (max 3 at a time)
- Apply answers atomically to spec.md
- Cite repo precedents for recommendations
Key principle: Planning is 80% of success — thorough clarification prevents costly rework.
Workflow position: spec → clarify → plan → tasks → implement → optimize → ship
Anti-Hallucination Rules
CRITICAL: Follow these rules to prevent fabricating ambiguities or solutions.
-
Never invent ambiguities not present in
spec.md.- ❌ BAD: "The spec doesn't mention how to handle edge cases" (without reading it)
- ✅ GOOD: Read spec.md, quote ambiguous sections: "spec.md:45 says 'users can edit' but doesn't specify edit permissions"
- Quote verbatim with line numbers:
spec.md:120-125: '[exact quote]'
-
Always quote the unclear text and cite line numbers for every question.
- When flagging ambiguity:
spec.md:120-125: '[exact quote]' - unclear whether this means X or Y - Don't paraphrase unclear text - show it verbatim
- Cite line numbers for all ambiguities
- When flagging ambiguity:
-
Never invent "best practice" without evidence.
- Don't say "Best practice is..." without evidence
- Source recommendations: "Similar feature in specs/002-auth used JWT per plan.md:45"
- If no precedent exists, say: "No existing pattern found, recommend researching..."
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 · 801 lines · 24 tokens per session scan A 31b5912db13b
clarify is a command published in the GitHub repository marcusgoll/Spec-Flow (91 stars, last pushed 4mo ago), licensed MIT. It adds 24 tokens to every session and 6,021 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 commands, from other repositories
sync-linear
Sync current work with Linear ticket status.
review
Review the current diff for violations of the conventions in AGENTS.md and .cursor/rules/. Report findings ordered by severity; do not fix anything unless asked.
auto-run
PitWay: Manage auto-run authorization for automatic task continuation.
task-integrate
PitWay: Apply a dispatched task's worktree commit to the main tree.
milestone-list
PitWay: List every milestone with a concise one-line status.
setup
Install the claude-pace statusline bundled with this plugin.