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 ychampion/cskill-agents --skill remote-safe-command-allowlistgit clone --depth 1 https://github.com/ychampion/cskill-agentsWrote 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/ychampion/cskill-agents/remote-safe-command-allowlist)<a href="https://agentmods.dev/skills/ychampion/cskill-agents/remote-safe-command-allowlist"><img src="https://agentmods.dev/badge/skills/ychampion/cskill-agents/remote-safe-command-allowlist/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/ychampion/cskill-agents/remote-safe-command-allowlist"><img src="https://agentmods.dev/badge/skills/ychampion/cskill-agents/remote-safe-command-allowlist.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 | $0.00039 | $0.00577 |
| Opus 5 | $0.00019 | $0.00289 |
| Sonnet 5 | $0.00008 | $0.00115 |
| Haiku 4.5 | $0.00004 | $0.00058 |
Grade A, and why
remote-safe-command-allowlist 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 7d 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 — 30 lines — stays where its author put it; the contents beside it link to each section on GitHub.
SKILL: Remote Safe Command Allowlist
Domain: command-architecture Trigger: Use when a remote session can expose a subset of local commands, but only commands with no filesystem, shell, IDE, or other host-bound dependencies should remain available. Source Pattern: Distilled from reviewed permission, shell-safety, and worktree-management implementations.
Core Method
Define one explicit allowlist for commands that are safe in remote mode because they only affect local TUI state or emit benign local information. Apply that allowlist before the remote REPL first renders so unsafe commands never flash into the initial command surface during handshake. Then reuse the same allowlist when the remote backend later narrows the advertised command list, preserving the locally safe commands alongside the server-provided set. This gives remote sessions one stable safety contract instead of separate startup and post-init rules that can drift apart.
Key Rules
- Keep remote safety as an explicit reviewed allowlist; default every other local command to blocked unless it proves it has no host-bound side effects.
- Reserve the allowlist for commands that only touch local UI state or similarly benign local behavior, not commands that need filesystem, git, shell, IDE, MCP, or other workstation context.
- Reuse the same allowlist at every remote-surface boundary, including initial pre-render filtering and later handshake refinement, so the visible command set stays consistent.
- Filter before the remote UI renders, not only after the backend init arrives; preventing a transient unsafe surface is part of the method.
- Preserve locally safe commands even when the remote backend returns its own command set, because some commands are safe precisely due to being local-only helpers.
Example Application
If a coding agent launches a mobile companion mode, create a remote safe commands set containing only commands like clear-screen, help, theme switching, or copying the last response. Filter the command palette through that set before the mobile UI appears, then when the remote server reports its own supported commands, merge that list with the same reviewed local-safe set so the user keeps harmless local controls without ever seeing unsafe host-bound actions.
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.
- 7d ago First seen · 30 lines · 39 tokens per session scan A 8a85c1ec624f
remote-safe-command-allowlist is a skill published in the GitHub repository ychampion/cskill-agents (36 stars, last pushed 5mo ago), licensed MIT. It adds 39 tokens to every session and 577 once invoked, about $0.0002 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 2026-09-03.
Other skills, from other repositories
grape
Use Grape MCP for Codex context continuity in coding repositories. Use when a task needs repeated-turn context, omitted context restore, stale-context checks, invalidation checks, or safe continuity across branch and dirty-worktree changes.
review
Validate plans, execution, or PRs against wish criteria — returns SHIP / FIX-FIRST / BLOCKED with severity-tagged gaps.
work
Execute an approved wish plan — orchestrate subagents per task group with fix loops, validation, and review handoff.
brainstorm
Explore ambiguous or early-stage ideas interactively — tracks wish-readiness and crystallizes into a design for wish.
genie
Entry point for Genie operations — routes bug reports, questions, and operational commands, resumes existing lifecycle state, and orchestrates work that needs durable planning or coordination. Other ordinary requests bypass the lifecycle with a one-line notice unless the user asks for Genie.
wish
Convert an idea into a structured wish plan with scope, acceptance criteria, and execution groups for work.