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.
git 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/start)<a href="https://agentmods.dev/commands/dwarvesf/dwarves-kit/start"><img src="https://agentmods.dev/badge/commands/dwarvesf/dwarves-kit/start/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/commands/dwarvesf/dwarves-kit/start"><img src="https://agentmods.dev/badge/commands/dwarvesf/dwarves-kit/start.svg" alt="Reviewed on agentmods" width="80" 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 5.1 | $0.00017 | $0.03118 |
| Opus 5.5 | $0.00007 | $0.01247 |
| Sonnet 5.5 | $0.00003 | $0.00624 |
| Haiku 4.5 | $0.00002 | $0.00312 |
Grade A, and why
start 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 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.
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 — 162 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Self-intro (AGENTS.md "Self-intro" convention): open your first reply with exactly one banner line, [kit:start] Detect project state and suggest the right next command., then the output for the selected mode. Exception: in --brief mode skip the banner; its one-line contract wins.
You are a project state detector. If _meta/BACKLOG.md exists, bash lib/board/backlog.sh board renders the kanban summary (mention queued count + suggest /kit:assign --next when items are waiting). If any run ledgers exist, bash lib/telemetry/lane-telemetry.sh misfires surfaces routing misfires worth a retro. Read the current project and suggest what the user should do next. Do NOT execute anything. Just detect and recommend.
Output mode (read $ARGUMENTS)
$ARGUMENTS selects how much to render. The detection Process below runs identically in every mode; only the output differs.
--brief-- emit exactly ONE line, max 120 characters: the matched state plus the single suggested next command, then the branch/dirty/spec tail in brackets. Nothing else. Example:Spec VALIDATED, 3/8 tasks -> /kit:execute. [master | 2 dirty | VALIDATED]- (no argument) -- DEFAULT. Emit the standard 3-4 line orientation defined in "Output format" below. This default output is unchanged from prior versions; do not alter it.
--full-- emit the default block, then append the "Full report" extras defined near the end of this file.
Process
Resolving the active spec (the dual-mode rule the hooks also use): among docs/specs/SPEC-*.md, the active spec is the lone non-SHIPPED/PARKED one; if several are live, the one whose slug matches the current git branch; if zero or multiple match the branch, the state is ambiguous, report spec:ambiguous(...) and ask which spec, never guess. docs/specs/SPEC-NNN-<slug>.md is the sole active-spec location (root only; co-located specs are found by spec-next, validate-round and ship-gate via lib/spec/spec-find.sh). States 3-8 below operate on that resolved active spec.
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 Changed · +1 lines 9e5368a4754c
- 14d ago Changed f3774b1e3a3b
- 16d ago Changed 0b791632bdd7
- 20d ago Changed 8aa84beedfc1
- 1mo ago Changed · +12 lines 752b2fdaadba
- 1mo ago Changed · +1 lines 8f2d7a3f63f9
- 1mo ago Changed be5d15e90446
- 1mo ago Changed · +1 lines c3f20503b409
- 1mo ago First seen · 147 lines · 17 tokens per session scan A 04cd49057e1f
start is a command published in the GitHub repository dwarvesf/dwarves-kit (11 stars, last pushed today), licensed MIT. It adds 17 tokens to every session and 3,118 once invoked, about $0.0001 per session on Opus 5.5. 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 2026-09-04.
Other commands, from other repositories
OPSX: Continue
Continue working on a change - create the next artifact (Experimental).
OPSX: Verify
Verify implementation matches change artifacts before archiving.
OPSX: Apply
Implement tasks from an OpenSpec change (Experimental).
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: Archive
Archive a completed change in the experimental workflow.