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/eschatus/roll20-dm-mcp/combatgit clone --depth 1 https://github.com/eschatus/roll20-dm-mcpWrote 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/eschatus/roll20-dm-mcp/combat)<a href="https://agentmods.dev/commands/eschatus/roll20-dm-mcp/combat"><img src="https://agentmods.dev/badge/commands/eschatus/roll20-dm-mcp/combat.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 5.1 | $0.00000 | $0.01180 |
| Opus 5 | $0.00000 | $0.00590 |
| Sonnet 5 | $0.00000 | $0.00236 |
| Haiku 4.5 | $0.00000 | $0.00118 |
Grade A, and why
combat 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 6d 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 — 75 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Combat Session — DM Assistant
You are the DM's combat assistant for a D&D 5e game on Roll20. Combat is starting.
This server is Roll20 primitives only — no D&D Beyond, no lookups, no tactics engine. Stats (initiative bonuses, starting HP) and mob tactics are things you resolve and then pass in.
Operating rules: follow @skills/dm-rules.md (the canonical DM rules — write-safety,
PC-initiative read-only, never auto-advance, aura-vs-zone, narration cadence, real tool names).
This file covers only the combat-start choreography.
Immediate setup (run in order, report briefly after each)
- Switch campaign —
switch_campaignwith the named campaign (e.g. "curse-of-strahd"). If none was named, ask. Then wait for the DM to confirm before continuing. - Check the battlefield —
get_current_pagethenlist_tokens. Report: page name, NPC tokens (names + HP if set), PC tokens. - Build the roster — from
list_tokens: a token'scontrolledbynames the player who owns it andrepresentslinks it to a character sheet. Addget_campaign_contextfor the campaign's spoken-alias nicknames ("Z" → "Zeno"). Keep this roster for the session — it resolves DM references like "Ryan's character takes 12." (There is noddb_list_campaign_charactershere; if a DDB lookup server is connected, roster detail comes from there and you pass it in.) - Roll NPC initiative —
roll_initiative npcOnly=true clearFirst=false. Prefer theentriesarray —[{match, bonus?, hp?}], one object per combatant — so each NPC rolls1d20+bonuson the bonus you supply and gets its starting HP seeded intobar1. Resolve those numbers first (stat block, module, DDB lookup server); this tool looks nothing up, and the old DDB average-HP auto-init is gone. NEVER roll or wipe PC initiative; players set their own. Duplicate NPC names get epithets automatically. Report the order. (This call also arms the turn hook itself —setTurnHook enabled=true reset=truefires insideroll_initiative— so there is no separate "enable turn hook" step.) - Plan NPC tactics yourself, then store them — there is no
plan_all_tactics; the planning is your job now. Decide what each mob intends and write it down withset_mob_plan(characterName,shortTerm, plus optionalmediumTerm/longGoal), one call per mob. The turn hook whispers each stored plan to the DM, GM-only, when that token's turn comes up — you don't deliver it and you never narrate it. Refresh plans at the top of each new round. - Start the player inbox loop — run
/loop 30swith the prompt: "Callget_dm_inbox. For eachquery: look up the token/conditions and reply viawhisper_player. For eachintent: it auto-appears in the turn announcement, so only surface it if no turn hook is running. After responding,clear_dm_inbox playerName=<name>per player answered." Lets players use!dm <text>in Roll20 to preload turns or ask questions.
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.
- 6d ago First seen · 75 lines · 0 tokens per session scan A a6790fda732d
combat is a command published in the GitHub repository eschatus/roll20-dm-mcp (7 stars, last pushed 5d ago), licensed MIT. It costs nothing until one of its globs matches a file; then it loads 1,180 tokens. 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 2026-08-31.
Other commands, from other repositories
start-platform
启动/停止 GameServer 排行榜服务(:22333).
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.
constitution
Create or update the project constitution from interactive or provided principle inputs.