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 skills add t1djani/war-room --skill discover-rostergit clone --depth 1 https://github.com/t1djani/war-roomWrote 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/skills/t1djani/war-room/discover-roster)<a href="https://agentmods.dev/skills/t1djani/war-room/discover-roster"><img src="https://agentmods.dev/badge/skills/t1djani/war-room/discover-roster/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/skills/t1djani/war-room/discover-roster"><img src="https://agentmods.dev/badge/skills/t1djani/war-room/discover-roster.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 | $0.00081 | $0.00679 |
| Opus 5 | $0.00041 | $0.00340 |
| Sonnet 5 | $0.00016 | $0.00136 |
| Haiku 4.5 | $0.00008 | $0.00068 |
Grade A, and why
discover-roster 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 11d 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 — 35 lines — stays where its author put it; the contents beside it link to each section on GitHub.
discover-roster
The default war-room runs five generic officers with zero config — fast, no scan. This skill is the opt-in upgrade: it scans the project once, infers its real domains, and proposes a roster where each officer reads a genuine slice of this codebase instead of a generic frame. Run it only when the user asks; keep the default light.
The output is a proposal. Nothing is written until the user accepts.
Procedure
-
Reuse before you scan. If a
.servo/manifest.yamlexists, derive the roster from itsexperts(domains + source pointers) — no scan needed. This is the cheap path; prefer it. -
Otherwise, scan — once, shallowly. Read the project's shape, not its contents: top-level layout, the package/build manifest, the
docs/index, the main source directories. You are looking for distinct domains (data, API, frontend, infra, billing…), not every file. Stop once you can name 3–5 domains; do not read the whole tree. -
Map each domain to an officer — persona + real slice. For each domain, propose:
- a persona (the skin) — an evocative war-council title that fits the domain;
- the slice (the skeleton) — the real paths/docs that domain owns;
- what it optimizes and its signature question, derived from the domain.
Keep the rule that makes a roster worth having: two officers must not share a model of the problem. If two candidate domains would ask the same question, merge them. Cap at 5 substantive officers.
-
Do NOT seat the Marshal or the Tenth Man. war-room always seats those itself (synthesis and mandated dissent are structural, not domains). The roster file lists only the substantive officers.
-
Propose, then confirm. Show the user the proposed roster — personas, slices, signature questions — and ask them to accept or adjust. Surface anything ambiguous as a question (a domain you weren't sure about, two that might be one). Never write the file without an explicit OK.
-
Write
.war-room/roster.yamlin the shape ofexamples/roster.example.yaml. Confirm where it landed and that the next/war-roomrun will use it.
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.
- 11d ago First seen · 35 lines · 81 tokens per session scan A 043b68ab801a
discover-roster is a skill published in the GitHub repository t1djani/war-room (1 stars, last pushed 2mo ago), licensed MIT. It adds 81 tokens to every session and 679 once invoked, about $0.0004 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 2026-08-31.
Other skills, from other repositories
learn
Use when the user says /learn or asks to save this lesson.
building
Implementation skill for writing production code with TDD. Covers the RED-GREEN-REFACTOR cycle, false-RED detection, vertical slicing, scope escalation, test process discipline, and code generation patterns. Loaded by component-builder and bug-investigator.
diff-driven-docs
Use when a BUILD phase completes, a commit is staged, or a PR is about to be created, and the diff has not yet been reflected in documentation. Also use when the user says "update docs", "sync docs", "document this", or asks whether documentation is up to date.
exploration
Two-mode exploration skill: (1) design dialogue — turn rough ideas into validated designs through collaborative interview before planning; (2) spike — throwaway code answering ONE design question, deleted or absorbed, never shipped. Router invokes mode via dispatch context.
memory-and-handoff
Two-mode skill: (1) session memory — load/persist durable workflow state under .cc10x/ (activeContext, patterns, progress) so context survives compaction; (2) handoff package — portable, secrets-redacted export for a coworker, different tool, or fresh non-cc10x session.
plan-review-gate
Use after saving a non-trivial plan or decision RFC when a fail-closed feasibility, completeness, and alignment review must block execution.