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/dwarvesf/dwarves-kit/specgit clone --depth 1 https://github.com/dwarvesf/dwarves-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/dwarvesf/dwarves-kit/spec)<a href="https://agentmods.dev/commands/dwarvesf/dwarves-kit/spec"><img src="https://agentmods.dev/badge/commands/dwarvesf/dwarves-kit/spec.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 | $0.00021 | $0.03785 |
| Opus 5 | $0.00010 | $0.01893 |
| Sonnet 5 | $0.00004 | $0.00757 |
| Haiku 4.5 | $0.00002 | $0.00379 |
Grade C, and why
spec scanned grade C with 1 finding 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 today.
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.
Hidden instructionshighPrompt injection
Directives inside HTML comments, invisible characters or bidirectional overrides are read by the model and not by the person reviewing the file.
<!-- ID-454, the PRE-build twin of ID-395's post-build visual proof. A ticket that carries a picture (a diagram or a prototype) builds better than prose alone. Required (non-empty) for a `full`-lane spec; encouraged, not How it starts
The opening of the file, as written. The whole thing — 284 lines — stays where its author put it; the contents beside it link to each section on GitHub.
You are a senior technical architect producing a development specification. The spec must be detailed enough for a contractor with no prior context to implement the feature correctly using Claude Code.
Process
Bracket the phase for timing (SPEC-129) before starting: bash lib/gate/gate-ledger.sh outcome <rid> Spec start.
Step 1: Gather intent
If a docs/briefs/DECISION-BRIEF.md exists, read it first (it may include a Solution design appended by /kit:design; fold that into the spec's ## Solution. It may also include a ## Design section, from the same command, with a diagram + ADR link(s); fold that into the spec's own ## Design , ADR-0031 §1). Otherwise, ask the user:
- What are you building? (one paragraph)
- Is this greenfield or modifying existing code?
- What's the tech stack? (or read from CLAUDE.md / package.json / go.mod)
- Who implements? (you, a contractor, a team)
Step 2: Research (if brownfield)
If modifying existing code, run codebase research before generating the spec. This keeps the main session's context clean.
Create docs/research/ directory first.
Mode A: Formal agents (preferred)
If the research agents are installed (check: do .claude/agents/research-stack.md etc. exist?), dispatch all 4 via the Task tool in parallel:
- research-stack agent: "Map the technology stack. Write to
docs/research/stack.md." - research-context agent: "Map existing features related to [user's feature area]. Write to
docs/research/features.md." - research-architecture agent: "Map architecture patterns and conventions. Write to
docs/research/architecture.md." - research-pitfalls agent: "Find landmines in [target area / target files]. Write to
docs/research/pitfalls.md."
Mode B: Inline fallback
If the formal agents are NOT installed, dispatch 4 Task tool subagents with these inline prompts:
Stack research:
Map the technology stack. Read package.json / go.mod / Cargo.toml / pyproject.toml and config files. Report: languages, frameworks, versions, key dependencies (top 5-10), build/test/deploy commands. If codebase-memory-mcp is available, use get_architecture(). Max 50 lines. Write to docs/research/stack.md.
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.
- today First seen · 284 lines · 21 tokens per session scan C b65a16c5178d
spec is a command published in the GitHub repository dwarvesf/dwarves-kit (11 stars, last pushed today), licensed MIT. It adds 21 tokens to every session and 3,785 once invoked, about $0.0001 per session on Opus 5. A static security scan graded it C with 1 finding (hidden instructions). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-09-04.
Other commands, from other repositories
OPSX: Archive
Archive a completed change in the experimental workflow.
OPSX: Explore
Enter explore mode - think through ideas, investigate problems, clarify requirements.
OPSX: Propose
Propose a new change - create it and generate all artifacts in one step.
OPSX: Sync
Sync delta specs from a change to main specs.
OPSX: Apply
Implement tasks from an OpenSpec change (Experimental).
OPSX: Update
Update a change - revise existing planning artifacts and keep them coherent (Experimental).