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 mblua/AgentsCommander --skill role-skill-boundary-auditgit clone --depth 1 https://github.com/mblua/AgentsCommanderWrote 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/mblua/agentscommander/role-skill-boundary-audit)<a href="https://agentmods.dev/skills/mblua/agentscommander/role-skill-boundary-audit"><img src="https://agentmods.dev/badge/skills/mblua/agentscommander/role-skill-boundary-audit/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/mblua/agentscommander/role-skill-boundary-audit"><img src="https://agentmods.dev/badge/skills/mblua/agentscommander/role-skill-boundary-audit.svg" alt="Reviewed on agentmods" width="80" height="20"></a>- NVIDIA SkillSpector pass
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 | $0.00046 | $0.00630 |
| Opus 5.5 | $0.00018 | $0.00252 |
| Sonnet 5.5 | $0.00009 | $0.00126 |
| Haiku 4.5 | $0.00005 | $0.00063 |
Grade A, and why
role-skill-boundary-audit 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 1mo 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 — 66 lines — stays where its author put it; the contents beside it link to each section on GitHub.
role-skill-boundary-audit
Purpose
Roles define who is responsible.
Skills define how to perform a reusable capability.
Diagnostic by default: recommend, do not rewrite, unless the user or active workflow asked for the refactor.
Conciseness mandate (always)
Write the minimum that changes behavior; roles and skills spend context budget every time they load. This applies to every Role.md and skill you write, recommend, or rewrite.
- Add only what adds value; cut the rest.
- Keep rationale to one line, and only where it guides a judgment the rule itself does not cover.
- No restatement: drop "Why" / "How to apply" / examples that repeat a rule without adding information.
Role.mdis the tightest surface (always loaded). Load-on-demand skills may hold more detail, but still no padding.- When recommending or rewriting, target the smallest, least-verbose change that restores the boundary and preserves operative meaning.
Classification
- Keep in Role: identity, ownership, authority, responsibilities, escalation, durable boundaries for one agent.
- Move to Skill: repeatable workflow, checklist, tool procedure, implementation pattern, domain method.
- Move to Global Policy: must constrain every agent or session regardless of role.
- Move to Workflow Docs: team process, operator/onboarding guide, durable docs humans browse outside startup context.
- Move to Memory: project fact, decision, preference, or status that persists but is not a standing instruction.
- Duplicate / Consolidate: same guidance in multiple places; pick one source of truth.
- Trim / Compress: content is in the right place but bloated; cut to the operative minimum.
- Split Agent: one role owns unrelated accountability surfaces.
- Merge Agent: agents differ mostly by wording or minor task variants.
- Needs Owner Decision: placement touches authority, access, team structure, or policy.
Workflow
- Identify the instruction or proposed change.
- Name current vs proposed location.
- Classify it.
- Check for authority language, reusable procedure, duplicated guidance, agent-boundary drift, and verbosity.
- Recommend the smallest, least-verbose change that restores a clear boundary and preserves meaning.
- Stop at the recommendation if the change would rewrite files or split/merge agents, unless the user asked for the refactor.
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.
- 1mo ago First seen · 66 lines · 46 tokens per session scan A af3e86a69e9a
role-skill-boundary-audit is a skill published in the GitHub repository mblua/AgentsCommander (12 stars, last pushed today), licensed MIT. It adds 46 tokens to every session and 630 once invoked, about $0.0002 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 2026-08-31.
Other skills, from other repositories
sivtr-memory
Retrieve shared local work memory: terminal activity, AI conversation history, prior decisions, validation evidence, debugging trails, recaps, and handoff context. Use before asking the user to repeat local work context.
lessons
How to propose and query distilled lessons (the experiential memory). Only project-steward persists; every other agent proposes/queries. Use when you discover a durable lesson or need prior experience for the current task.
worth-fix
A verification and decision-making guide for checking whether bug reports, claims, proposals, or refactoring ideas are true and worth acting on.
remem
Use when the user asks Codex to recall prior project context, save durable decisions or bug fixes, inspect remem memory health, or activate remem automatic memory hooks from the Codex plugin.
performance-checklist
Performance review checklist — hot paths, async-runtime discipline, streaming, query-client tuning, layout pre-measurement, token/cost. Load when reviewing perf-sensitive changes.
moli-webfetch
Fetch, inspect, crawl, and capture live, JavaScript-rendered websites with Moli. Use when Codex needs current web content, web research, fact lookup, link following, a bounded crawl, client-rendered or response-gated content, network diagnostics, or a standalone HTML, Markdown, JSON, semantic-tree, viewport or…