codex-app-server-product-integration
Skill0 starsChanged 3 months agoApache-2.0View on GitHub
Use when designing or updating an agent-enabled local app, GUI, Mac app, IDE-like client, image-generation GUI, personal agent, code-review app, or any product that should drive Codex through codex app-server, JSON-RPC/JSONL, approval loops, streaming events, or ChatGPT subscription-backed local Codex auth.
In this repository78 other files it gives its agents.
- AGENTS.mdAGENTS.md
- .agents/skills/go-skills-knowledge-pack/AGENTS.mdAGENTS.md
- .agents/skills/rust-skills-knowledge-pack/AGENTS.mdAGENTS.md
- .agents/skills/typescript-skills-knowledge-pack/AGENTS.mdAGENTS.md
- .claude/skills/go-skills-knowledge-pack/AGENTS.mdAGENTS.md
- .claude/skills/rust-skills-knowledge-pack/AGENTS.mdAGENTS.md
- .claude/skills/typescript-skills-knowledge-pack/AGENTS.mdAGENTS.md
- CLAUDE.mdCLAUDE.md
- .cursor/rules/00-general.mdcCursor rule
- .cursor/rules/revharness-critical.mdcCursor rule
- .cursor/rules/revharness-detailed.mdcCursor rule
- auto-orchestratorSkill
- baseline-protectionSkill
- client-distribution-readinessSkill
- cloudflare-deploy-guardSkill
- codex-app-server-guardSkill
- codex-callerSkill
- context-compactorSkill
- cursor-callerSkill
- delegated-development-disciplineSkill
- deploy-command-disciplineSkill
- design-principlesSkill
- development-constitutionSkill
- development-junk-cleanupSkill
- go-skills-knowledge-packSkill
- harness-official-docs-updateSkill
- naming-normalization-guardSkill
- orchestrator-bootstrapSkill
- payload-cms-deploy-guardSkill
- production-function-implementerSkill
- research-handoffSkill
- rev-harness-lifecycleSkill
- review-workflowSkill
- root-cause-first-disciplineSkill
- rust-skills-knowledge-packSkill
- scope-guardSkill
- self-growth-proposal-triageSkill
- shared-client-wrapper-safetySkill
- slice-designerSkill
- staff-code-reviewerSkill
- structured-mentorSkill
- supabase-deploy-guardSkill
- system-plannerSkill
- typescript-skills-knowledge-packSkill
- auto-orchestratorSkill
- baseline-protectionSkill
- client-distribution-readinessSkill
- cloudflare-deploy-guardSkill
- codex-app-server-guardSkill
- codex-app-server-product-integrationSkill
- codex-callerSkill
- context-compactorSkill
- cursor-callerSkill
- delegated-development-disciplineSkill
- deploy-command-disciplineSkill
- design-principlesSkill
- development-constitutionSkill
- development-junk-cleanupSkill
- go-skills-knowledge-packSkill
- harness-official-docs-updateSkill
- The first 60 shown.
The file
We split this skill into the rules it gives an agent
4 rules. About 0 are common, 1 is distinctive, and 3 are very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈1
Distinctive · estimate
This author's own choices. Worth a look.
3
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Very specific: line 8
Use this skill before choosing SDK, direct API, codex exec, wrapper automation, or Claude Code plugin integration for an agent-enabled …
- Very specific: line 19
Prefer stdio:// for embedded local clients.
- Very specific: line 20
Treat ws://IP:PORT as experimental.
- Distinctive: line 22
Do not expose a user's local Codex subscription over an unauthenticated network listener.
≈ “Common” is an estimate: at least 1 in 5 of the 364,980 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
Codex App Server Product Integration
Use this skill before choosing SDK, direct API, codex exec, wrapper automation, or Claude Code plugin integration for an agent-enabled application.
Decision Rule
- Choose
codex app-serverfirst for local apps that need a long-lived Codex agent, bidirectional streaming, multi-turn state, approval round-trips, file/shell/image/search/MCP tool access, or a GUI over Codex. - Choose
scripts/codex-wrapper.sh --role ...orcodex execfor deterministic one-shot harness automation, CI, cron, release-gate checks, or reviewer/coder calls. - Choose the OpenAI SDK or API when the application server owns auth, billing, typed API calls, or non-Codex product behavior.
- Choose OpenAI
codex-plugin-ccwhen Claude Code is the operator UI and the user wants official Claude Code -> Codex review/delegation UX.
Transport Defaults
- Prefer
stdio://for embedded local clients. It is the defaultcodex app-servertransport and uses newline-delimited JSON. - Treat
ws://IP:PORTas experimental. Use loopback only unless a reviewed design adds WebSocket auth and network exposure controls. - For WebSocket, require token or signed bearer auth before any non-loopback exposure.
- Do not expose a user's local Codex subscription over an unauthenticated network listener.
Revharness Boundaries
- App-server is a product integration transport, not acceptance truth.
- Goal/app-server state must not replace ExecPlan, SOW, task lineage ledger, deterministic checks, reviewer LGTM, or completion boundary.
- Subscription-only remains the default: do not introduce API-key dependency for this workflow unless the user explicitly changes the policy.
- Local trust is the security model. The authenticated Codex user is the local machine user behind
codex login. - Shell/file write capability must be surfaced in the product UX with explicit trust and approval boundaries.
Minimal Checks Before Implementation
codex app-server --help
rg -n "OpenAI Codex App Server|OpenAI Codex Plugin for Claude Code" docs/official-docs-links.mdFor Revharness behavior changes, route through harness-official-docs-update and record official docs consulted, local authority mapping, and deterministic checks before review.
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
No reviews yet
People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.
- It worked
- Partly
- It didn't
Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

