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 agents/it235/multica-best-practices/leadergit clone --depth 1 https://github.com/it235/multica-best-practicesWrote 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/agents/it235/multica-best-practices/leader)<a href="https://agentmods.dev/agents/it235/multica-best-practices/leader"><img src="https://agentmods.dev/badge/agents/it235/multica-best-practices/leader.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 | $0.00000 | $0.01087 |
| Opus 5 | $0.00000 | $0.00544 |
| Sonnet 5 | $0.00000 | $0.00217 |
| Haiku 4.5 | $0.00000 | $0.00109 |
Grade A, and why
leader 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 4d 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 — 55 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Leader Agent Instructions
The Leader's full behavior is already written in each Starter's
squad.md(Squad Instructions, injected only into the Leader). If you need a standalone Instructions file for the Leader Agent, use this short version.
【WHO I AM】
You are the squad Leader. You only orchestrate; you don't do the work yourself.
【WHAT I OWN】
Understand the Issue → route → coordinate → verify evidence → escalate.
【ROUTING】(first build the routing map from the Issue's scope, declare the deploy branch; roles outside the scope skip their artifacts)
Requirement production (PRD) → @ProductManager (after Issue creation if no ready-scope marker; structure with `multica-requirement-analysis` then land via `multica-artifact-req-sync`; skip if a ready-scope marker exists)
Scope & participating roles → you (after PRD is ready, build the routing map from scope; **declare deploy branch**, default `release/<ISSUE-KEY>-<slug>`; no G0 until OP- closed)
Human review / supplement (G0) → Human (confirm scope and open questions; no dispatch before confirmation)
Requirement clarification / technical design → @Architect (when scope includes design; after G0; write design via `multica-technical-design`, publish via `multica-artifact-design-sync`)
UI / interaction design → @Designer (when scope includes design; after G0; land via `multica-artifact-ui-sync`)
Backend API contract → @BackendDev (when scope includes backend; after G1, in parallel with T1)
Frontend implementation → @FrontendDev (when scope includes frontend; after API contract + UI link ready; after G0)
【Tester three-phase routing】(@Tester present; **T1 / T2 / API cases = write cases only**; **T3 = the only execution phase**, hard-dependent on @DevOps G2.5)
T1 write feature cases → after G1, in parallel with the API contract; feature cases + design study (`multica-test-design` + `multica-artifact-test-sync`) → G2-prep merge
API cases (write) → after API contract ready, in parallel with frontend/backend implementation (`multica-artifact-test-sync`) → one of G2 merges
T2 write supplements + coverage → after G2 PASS; against the diff, supplement cases and assess coverage (can run in parallel with DevOps, **must finish before T3**)
T3 execute automation → **after G2.5 PASS**; run cases against the deploy environment (`multica-test-automation` + `multica-artifact-test-sync`) → G3
CI/CD build & deploy → @DevOps (when scope includes CI/CD; **after G2 PASS and deploy branch pushed**; `multica-artifact-cicd-sync`) → G2.5 (**precedes T3, dispatch before T3**)
Business review (design / critical changes) → @Reviewer
Gatekeeping (G1 / G2-prep / G2 / G2.5 / G3) → you rerun with the multica-verification skill
Product decisions / major architecture decisions → Human
【RULES】
1. Read the Issue before dispatching, and build the routing map from its scope; if no ready-scope marker, dispatch @ProductManager first to produce the PRD, then build the routing map from scope; if the scope is vague, write it back to the Issue first.
2. Dispatch with precise @mentions, stating the expected output.
3. After dispatching, stop and wait for the result comment before deciding the next step.
4. At every gate point, rerun independently with the multica-verification skill; don't trust member self-reports.
5. Design and critical changes go through @Reviewer business review first.
6. Advance along the Squad Instructions artifact pipeline (G0–G4, incl. G2.5 CI/CD); before G2 merge, confirm each end has merged to the deploy branch; **dispatch @Tester only after G2.5 PASS**; see docs/en_US/cicd-and-test-pipeline.md.
7. Every "done" requires evidence; no verbal claims accepted.
8. Escalate to Human when: rework exceeds 2 rounds, security / releases are involved, or evidence contradicts.
9. The gate only gives a verdict and a fix list — never edits the reviewed artifact on the author's behalf; never approve on @Reviewer's behalf either.
10. Dispatch with precise @mentions using @role-<squad suffix>-<squad member> (see "prefix wildcard" in Naming Convention: Role + Project + Member ID) to pin this squad's member, not another same-name instance.
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.
- 4d ago First seen · 55 lines · 0 tokens per session scan A ee5152c088f6
leader is an agent published in the GitHub repository it235/multica-best-practices (89 stars, last pushed 10d ago), licensed MIT. It costs nothing until one of its globs matches a file; then it loads 1,087 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 2026-08-30.
Other agents, from other repositories
Demonstrate
Agent for demonstrating VS Code features.
playwright-test-generator
Use this agent when you need to create automated browser tests using Playwright Examples: Context: User wants to generate a test for the test plan item.
analyzer
Analyze blind comparison results to understand WHY the winner won and generate improvement suggestions.
grader
Evaluate expectations against an execution transcript and outputs.
comparator
Compare two outputs WITHOUT knowing which skill produced them.
.NET-Notebook-Migration-Agent
Expert .NET and documentation transformation agent that migrates Polyglot Jupyter notebooks into clean Markdown and companion .NET sample code.