setup
Skill2 starsChanged 5 months agoApache-2.0View on GitHub
Use to initialize the Volley Claude+Codex workflow in this repo. Runs once per repo. Verifies Codex is installed and can run non-interactively, scaffolds .volley/ with HANDOFF.md template, .gitignore, and an initial STATE file. Runs a smoke test that calls Codex via `codex exec`.
- Installs packages
191 other projects use a version of this.
Adopted by
Most recently taken up by
- GoogleCloudPlatform/gemini-cloud-assist-mcptheir own version, first seen Nov 15, 2026
- finolabs/vibecodetheir own version, first seen Nov 15, 2026
- quickchatai/quickreviewtheir own version, first seen Nov 15, 2026
- tomw1808/flash-agentstheir own version, first seen Nov 15, 2026
- felipeam86/pipe-skillstheir own version, first seen Nov 15, 2026
- oss-invisible/playwright_mcptheir own version, first seen Nov 15, 2026
Used with
Projects with it also use add-voice-transcription, add-discord and 10 more.
Other files people use alongside this one.
Files found in the same public projects as a version of this one, more often than they turn up anywhere else. Each project counts once, large collections count for less, and this file's own repository is left out.
juancruzmunozalbeloSkilladd-voice-transcription
Add voice message transcription to NanoClaw using OpenAI's Whisper API. Automatically transcribes WhatsApp voice notes so the agent can read and respond to them.
Also in 5 projects that use this
MicrosoftSkilladd-discord
Add Discord bot channel integration via Chat SDK.
Also in 11 projects that use this
MicrosoftSkilladd-slack
Add Slack channel integration via Chat SDK.
Also in 11 projects that use this
juancruzmunozalbeloSkilladd-telegram-swarm
Add Agent Swarm (Teams) support to Telegram. Each subagent gets its own bot identity in the group. Requires Telegram channel to be set up first (use /add-telegram). Triggers on "agent swarm", "agent teams telegram", "telegram swarm", "bot pool".
Also in 7 projects that use this
- +8 more from microsoft
Other versions191 projects use a version of this.
Word for word · 0
None found.
Edited · 0
None found.
Wrote their own version · 191
- anthropics/knowledge-work-plugins★ 36ksame name
- garrytan/gbrain★ 32ksame name
- nanocoai/nanoclaw★ 31ksame name
- qwibitai/nanoclaw★ 31ksame name
- gavrielc/nanoclaw★ 31ksame name
- Q00/ouroboros★ 6.4ksame name
- pytorch/executorch★ 5.2ksame name
- aws/agent-toolkit-for-aws★ 3.5ksame name
- agenticnotetaking/arscontexta★ 3.5ksame name
- spotify/portal-ai-plugins★ 2.8ksame name
181 more not listed
How matches are found. Word for word: the exact same text. Edited: shares passages with this file (“99% same” means 99% of its five-word phrases also appear here; 10% or more counts). Their own version: similar meaning, or, weakest, the same name. We can’t tell which project had it first. Projects are counted among the 200 most-starred candidates, and mentions are links to the repository we found, up to 50 per site, so both counts are a floor.
In this repository7 other files it gives its agents.
- diagnoseSkill
- implementSkill
- review-codeSkill
- review-planSkill
- review-prSkill
- statusSkill
- unlockSkill
The file
We split this skill into the rules it gives an agent
11 rules. 0 are common, 0 are distinctive, and 11 are very specific to this project.
0
Common
Most of the 170 edited versions we read keep these.
0
Distinctive
This author's own choices. Worth a look.
11
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Very specific: line 13
Run: codex --version — must succeed and print a version >= 0.156.
- Very specific: line 17
Run codex exec --help - it must succeed.
- Very specific: line 18
If this repo has a project .mcp.json from an older Volley (or the user added one) that registers codex mcp-server, tell the user that entry …
- Very specific: line 30
Write machine-local state .volley/local.json with this repo's identity (gitignored - never committed).
- Very specific: line 45
Read .volley/config.json (parse it yourself - it is JSON, no jq needed).
- Very specific: line 57
Do not save the smoke test's SESSION_ID to any role.
- Very specific: line 58
This doubles as model validation (no hard-coded allowlist): If Codex returns a "model not found / unavailable" style error, report it …
- Very specific: line 59
If the call fails with an auth error, tell the user to run codex login and retry.
- Very specific: line 62
Use volley_next_step from the lib:
- Very specific: line 79
Model/effort validation: never maintain a hard-coded model list.
- Very specific: line 84
Codex not logged in: Run codex login (interactive — user must do this themselves).
“Common” is measured: kept by at least 6 in 10 of the 170 edited versions of this file we could read. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
/volley:setup
One-time installation of the Volley workflow.
Steps for Claude
- Verify Codex is installed.
- Run:
codex --version— must succeed and print a version >= 0.156. - We do NOT run an explicit
codex loginprobe here; the exact auth-status subcommand varies by Codex CLI version and a missing/wrong subcommand could falsely report "not logged in." Instead, the smoke test in step 4 will surface auth errors clearly when it calls Codex. If the smoke test fails with an auth error, instruct the user to runcodex loginand retry.
- Run:
- Confirm Codex can run non-interactively.
- Run
codex exec --help- it must succeed. Volley calls Codex throughscripts/codex-exec.sh(codex exec); it no longer bundles an MCP server, because Codex 0.156 removedcodex mcp-server. - If this repo has a project
.mcp.jsonfrom an older Volley (or the user added one) that registerscodex mcp-server, tell the user that entry is dead and can be removed. Do not edit it yourself.
- Run
- Scaffold
.volley/.- Create directory:
mkdir -p .volley - Copy template:
cp "${CLAUDE_PLUGIN_ROOT}/scripts/templates/HANDOFF.md" .volley/HANDOFF.md(only if.volley/HANDOFF.mddoesn't already exist - never overwrite) - Copy the managed checkpoint:
cp "${CLAUDE_PLUGIN_ROOT}/scripts/templates/CHECKPOINT.md" .volley/CHECKPOINT.md(only if it doesn't already exist - never overwrite; this is the Volley-managed cross-agent state file, distinct from the user-owned HANDOFF.md). - Copy the project config only if absent (never overwrite - the user may have customized it):
cp "${CLAUDE_PLUGIN_ROOT}/scripts/templates/config.json" .volley/config.json. An absentconfig.jsonmeans "today's default behavior"; the scaffolded defaults (model: inherit, no required context files) reproduce that. - Copy gitignore:
cp "${CLAUDE_PLUGIN_ROOT}/scripts/templates/gitignore" .volley/.gitignore(overwrite OK; this is internal). - Initialize STATE:
. "${CLAUDE_PLUGIN_ROOT}/scripts/lib.sh" && volley_state_write .volley/STATE claude idle 0 - Write machine-local state
.volley/local.jsonwith this repo's identity (gitignored - never committed). Derive identity with the lib helpers and write the JSON yourself (host-parsed - no jq):
Then create. "${CLAUDE_PLUGIN_ROOT}/scripts/lib.sh" ROOT=$(volley_repo_root); REMOTE=$(volley_repo_remote) echo "root=$ROOT remote=$REMOTE".volley/local.json(only if absent) as:{ "version": 1, "repository": { "canonicalRoot": "<ROOT>", "remote": "<REMOTE>" }, "modelOverrides": { "review": null, "implementation": null }, "roles": {} }
- Create directory:
- Smoke-test Codex - and validate the configured model against the live surface.
- Read
.volley/config.json(parse it yourself - it is JSON, no jq needed). Resolve the review role'smodelandreasoningEffort. - Run:
Do not save the smoke test's. "${CLAUDE_PLUGIN_ROOT}/scripts/lib.sh" printf 'Reply with the single word: PONG\n' > .volley/smoke-prompt.md if bash "${CLAUDE_PLUGIN_ROOT}/scripts/codex-exec.sh" --prompt-file .volley/smoke-prompt.md --out .volley/smoke-raw.md \ --cwd "$(volley_repo_root)" --model "<model or inherit>" --effort "<effort or inherit>"; then grep -q PONG .volley/smoke-raw.md && echo "[PASS] Codex replied PONG" || echo "[FAIL] Codex replied, but not with PONG" else echo "[FAIL] codex-exec.sh exited $? - read the error above" fiSESSION_IDto any role. - This doubles as model validation (no hard-coded allowlist). If Codex returns a "model not found / unavailable" style error, report it verbatim and tell the user to fix
modelin.volley/config.json(or.volley/local.jsonoverrides) - never silently fall back to another model. - If the call fails with an auth error, tell the user to run
codex loginand retry.
- Read
- Print the next-step block.
- Use
volley_next_stepfrom the lib:. "${CLAUDE_PLUGIN_ROOT}/scripts/lib.sh" volley_next_step "/volley:status" "Verify the lock state and sanity-check the install."
- Use
What to do on already-installed repos
If .volley/STATE exists, skip STATE re-init, but still backfill any missing new-in-v0.2 files (never overwrite existing ones): .volley/config.json, .volley/CHECKPOINT.md, and .volley/local.json (with this repo's identity). Refresh .volley/.gitignore from the template. Then re-run the smoke test (step 4) and print the next-step block. This is the migration path for repos set up before model selection existed - existing behavior is preserved because the scaffolded config defaults to inherit.
Configuration reference
Two files, one shared and one local. Claude parses both natively (no jq).
.volley/config.json(shared, committed):version(integer), acodexblock withreviewandimplementationroles (model: a concrete id or"inherit";reasoningEffort: e.g.high/xhighor"inherit";continuity:session-only|rehydrate|resume-if-safe), and acontextblock (requiredfiles that must exist or setup/review fails;optionalfiles reported-and-skipped if absent;managedCheckpointpath, default.volley/CHECKPOINT.md). Absent file = today's defaults. (The separate configurable-roles feature adds arolesblock under the sameversion- leave it to that feature; don't scaffold it here.).volley/local.json(machine-local, gitignored):repositoryidentity (canonicalRoot,remote) used to guard against cross-project resume;modelOverridesfor this machine; androlesholding exactthreadId/sessionIdper role. Never committed - IDs are per-user/per-machine.
Model/effort validation: never maintain a hard-coded model list. The step-4 smoke call is the validation - an unavailable model surfaces as a real Codex error. Reasoning effort rides config.model_reasoning_effort (there is no dedicated field).
Failure modes
- Codex not installed: Tell user to install:
npm install -g @openai/codex(or whatever their install path was). - Codex not logged in: Run
codex login(interactive — user must do this themselves). - Errors mentioning
codex mcp-serveror MCP: an old Volley-era.mcp.jsonis still registered; remove that entry - Volley no longer uses MCP.
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.

