cross-provider-agent
Dispatch a bounded task to a Claude, Codex, Cursor, or OpenCode CLI agent with least authority.
Skill by ejarmand
Chalk Labs
Need help deploying cross-provider-agent?
Chalk Labs sets up agents like this inside real teams: connected to your tools, tested on your data, with a human in the loop where it matters.
Install
Works with Claude Code, Claude apps, Codex, Cursor, Gemini CLI, GitHub Copilot, VS Code, OpenCode, Goose, Windsurf. Read the repository before you give any tool access to your data.
npx agentmarket install skill/ejarmand-cross-provider-agent
Asks which of your tools to add it to (Claude Code, Codex, Cursor, VS Code, Windsurf, Gemini CLI, Claude Desktop) and shows the change first. See https://agentmart.dev/cli.
Get the weekly diff
The weekly diff: one email a week with new agent tools, notable updates and grade changes, plus security alerts for tools you follow. Double opt-in, one-click unsubscribe.
About
Use cases: productivity, agent-building
Reviews
Write a reviewNo reviews yet. Installed it? Reviews come from verified installers only. How reviews work
Permissions
What this skill can do and what it asks for, from its SKILL.md. Checked Oct 25, 2026.
- Tools
- No tools found
- Secrets
- None asked for
- Install scripts
- None found
Network hosts it talks to 1
codeload.github.com· found in install command
Files it touches
~/.claude/skills/cross-provider-agent
What's inside
- Skill name
cross-provider-agent- Description
- Dispatch a bounded task to a Claude, Codex, Cursor, or OpenCode CLI agent with least authority. Use for independent reviews, named authority profiles, or external-provider subagents.
- License
- MIT
- Instructions
- 39 lines in SKILL.md · view on GitHub
Files (3)
- SKILL.md
- openai.yaml · agents/
- github-pr-reviewer.md · profiles/
SKILL.md (first lines)
# Cross-Provider Agent
Use a subagent through a cli. Per cli invocations are under skills:
`/claude-agent`
`/codex-agent`
`/cursor-agent`
`/opencode-agent`
## Choosing harness/model provider
1. For models with a particular provider plan, always use their native harness and subsidized plan
- e.g. codex - gpt models, claude code - anthropic models, cursor - grok models
2. Deffering to 1 or user instructions, prefer spwaning subagents using the harnesses own subagent tool
3. When independence matters (e.g. reviewing or judging work) exclude the
model provider that produced the work.
## Authority profiles
Profiles are provider-neutral authority contracts applied by adapters to one
fresh child. Adapter encodings live under `profiles/<name>/` beside each
adapter's SKILL.md.
### Profile index
- [`github-pr-reviewer`](profiles/github-pr-reviewer.md) — read and enforce
before choosing an adapter.
## Workspace ownership
Require an absolute path to a caller-prepared workspace. The caller owns
checkout creation, pinning, verification, and deletion.
## Distinguish network denial from authentication failure
Transport errors (`dial`, `lookup`, `connect`, loopback failures) indicate
sandboxed network access; an HTTP 401 "Bad credentials" response indicates
authentication failure.From ejarmand/skills, MIT.
More from ejarmand/skills
- 0.592score
cursor-agent
ejarmand
Run Cursor Agent CLI as an independent coding agent for implementation, investigation, planning, or review in a local workspace.
SkillAFreshness: A★ 1Claude Code +9 - 0.589score
opencode-agent
ejarmand
Run OpenCode CLI as an independent coding agent across its provider catalog, capture its session, or dispatch its profiled GitHub PR reviewer.
SkillAFreshness: A★ 1Claude Code +9 - 0.584score
pr-ping-pong
ejarmand
Drive an issue or pull request through alternating rallies of implementation and cross-provider review until it is review-clean.
SkillAFreshness: A★ 1Claude Code +9 - 0.581score
file-pr
ejarmand
Structure GitHub pull requests, issues, and comments.
SkillAFreshness: A★ 1Claude Code +9 - 0.562score
code-review
ejarmand
Review the changes since a fixed point (commit, branch, tag, or merge-base) along two axes — Standards (does the code follow this repo's documented coding…
SkillAFreshness: A★ 1Claude Code +9 - 0.559score
improve-codebase-architecture
ejarmand
Scan a codebase for deepening opportunities, present them as a visual HTML report, then grill through whichever one you pick.
SkillAFreshness: A★ 1Claude Code +9 - 0.552score
fix-steering
ejarmand
Analyze Codex history for interruptions, corrections, and user steering that indicate agent errors.
SkillBFreshness: B★ 1Claude Code +9 - 0.552score
codex-agent
ejarmand
Run Codex CLI as an independent coding worker for implementation, investigation, planning, or review, then monitor and verify its result.
SkillBFreshness: B★ 1Claude Code +9 - 0.547score
claude-agent
ejarmand
Run Claude Code CLI headless as an independent coding agent for implementation, investigation, planning, or review in a local workspace.
SkillBFreshness: B★ 1Claude Code +9 - 0.547score
codebase-design
ejarmand
Shared vocabulary for designing deep modules, and home of the unified code standard (STANDARD.md).
SkillBFreshness: B★ 1Claude Code +9 - 0.547score
html-visualization
ejarmand
Build, revise, and evaluate browser-rendered HTML visualizations for concepts, systems, codebases, architectures, workflows, timelines, comparisons, and other…
SkillBFreshness: B★ 1Claude Code +9 - 0.542score
implement
ejarmand
Implement a prepared spec or set of tickets.
SkillBFreshness: B★ 1Claude Code +9
Signals
- Stars
- 1
- Forks
- 0
- Open issues
- 20
- Last commit (this folder)
- Sep 22, 2026 (1 month ago)
From GitHub (fetched Oct 24, 2026).
Details
- License
- MIT
- Repository
- github.com/ejarmand/skills
- First seen
- Oct 24, 2026
- Last checked
- Oct 24, 2026
- Source
- GitHubgithub-graphql
More from ejarmand
- 0.592score
cursor-agent
ejarmand
Run Cursor Agent CLI as an independent coding agent for implementation, investigation, planning, or review in a local workspace.
SkillAFreshness: A★ 1Claude Code +9 - 0.589score
opencode-agent
ejarmand
Run OpenCode CLI as an independent coding agent across its provider catalog, capture its session, or dispatch its profiled GitHub PR reviewer.
SkillAFreshness: A★ 1Claude Code +9 - 0.584score
pr-ping-pong
ejarmand
Drive an issue or pull request through alternating rallies of implementation and cross-provider review until it is review-clean.
SkillAFreshness: A★ 1Claude Code +9 - 0.581score
file-pr
ejarmand
Structure GitHub pull requests, issues, and comments.
SkillAFreshness: A★ 1Claude Code +9 - 0.562score
code-review
ejarmand
Review the changes since a fixed point (commit, branch, tag, or merge-base) along two axes — Standards (does the code follow this repo's documented coding…
SkillAFreshness: A★ 1Claude Code +9 - 0.559score
improve-codebase-architecture
ejarmand
Scan a codebase for deepening opportunities, present them as a visual HTML report, then grill through whichever one you pick.
SkillAFreshness: A★ 1Claude Code +9
Done for you
Rather not set it up yourself?
Chalk Labs scopes, builds and runs agents for your team: the task audit, the eval harness, the integrations and the weekly ops report. Fixed scope, priced before we start.
See how we build AI agents →Read the case studies
Chalk Labs is a two-founder AI and growth studio. We build AI agents, CRMs and MVPs for Web3 and AI companies, and have shipped 50+ product and token launches.
- launches shipped
- 50+
- launches shipped
- media placements
- 500+
- media placements
- founders on every account
- 2
- founders on every account
Press we've placed: Forbes · CoinDesk · Decrypt · Cointelegraph · The Block
Rahil Jain, Co-founder, Chalk Labs. Ex-Goldman Sachs engineer. You talk to the person who builds it.
Get cross-provider-agent running in your business
Tell us what you want to automate. A founder replies within one working day with a plan, not a pitch deck.
Cite this page
cross-provider-agent skill. Agent Mart. https://agentmart.dev/skills/ejarmand-cross-provider-agent (last updated 2026-10-24)