mirror of
https://github.com/tiennm99/ccs.git
synced 2026-08-30 22:23:53 +00:00
Kai (Tam Nhu) TranandGitHub
9aa854bc59
* feat(docker): publish ccs:latest + ccs:full integrated images (P1)
* feat(docker): parameterize Dockerfile.integrated with ARG FLAVOR=minimal|full
Add FLAVOR build arg that gates the AI CLI install layer (claude-code,
gemini-cli, grok-cli, opencode) so one Dockerfile produces both the
minimal (< 350 MB) and full (< 600 MB) integrated images.
Use BuildKit cache mount for /root/.npm to speed up repeated builds.
Part (P1 — publish integrated images).
* feat(docker): emit startup deprecation warning in legacy ccs-dashboard entrypoint
Prepend a [WARN] line to stderr on every container start so operators
running ghcr.io/kaitranntt/ccs-dashboard:latest are notified to migrate
to ghcr.io/kaitranntt/ccs:latest. Sunset window: 2 releases. See .
* test(docker): add image-size.sh budget assertion + unit test suite
image-size.sh: asserts docker image inspect .Size against a byte budget.
Usage: image-size.sh <image:tag> <max-bytes>
Budgets: minimal=350 MB (367001600), full=600 MB (629145600)
image-size-logic.test.sh: 6 mock-docker unit tests covering pass/fail
boundaries, bad arg count, and non-integer input. All 6 pass locally.
Part (P1).
* ci(docker): extend docker-release.yml to publish ccs:latest and ccs:full
- Add publish-integrated job with matrix flavor: [minimal, full]
- minimal -> ghcr.io/kaitranntt/ccs:<ver> + :latest (when promoted)
- full -> ghcr.io/kaitranntt/ccs:full-<ver> + :full (when promoted)
- Multi-arch: linux/amd64 + linux/arm64 via buildx
- Scoped GHA cache per flavor to avoid cross-contamination
- Add promote_to_latest workflow_dispatch input (default false)
- rc.1 soak: first publish only pushes immutable version tag
- Mutable :latest/:full promoted only on release event OR explicit opt-in
- Add smoke-test job (post-publish) for each flavor
- Pulls the just-published version tag
- Asserts image size via tests/docker/image-size.sh
- Boots container, waits for healthcheck, probes :3000 and :8317
- Keep publish-dashboard job unchanged (legacy 2-release sunset)
- Updated labels to note deprecation status
All jobs run on self-hosted cliproxy runners. Part (P1).
* docs(docker): document new image tags, add ccs-dashboard deprecation notices
docker/README.md:
- Add "Choosing an image" table (ccs:latest / ccs:full / ccs-dashboard deprecated)
- Update Quick Start section to use ccs:latest as primary example
- Add legacy image note with sunset timeline
CHANGELOG.md:
- Add Unreleased > ### Deprecated entry for ccs-dashboard:latest
pointing to migration path and
README.md:
- Add one-line deprecation banner near top routing users to ccs:latest
Part (P1).
* refactor(docker): drop Bun from runtime stage; generate npm lockfile in build stage
- Build stage keeps Bun for fast installs; appends `npm install --package-lock-only`
to generate an ephemeral package-lock.json immediately after `bun run build:all`.
- Runtime stage removes BUN_VERSION ARG/ENV, the bun.sh curl install, and
`bun install --frozen-lockfile --production`. Replaces with:
COPY --from=build /app/package.json /app/package-lock.json ./
RUN npm ci --omit=dev --ignore-scripts
- --ignore-scripts rationale: postinstall (scripts/postinstall.js) writes ~/.ccs/
config — not needed in Docker context. bcrypt v6+ is pure-JS, no native compile.
- package-lock.json already in .gitignore (line 33); never committed.
- docker/Dockerfile.integrated unchanged — no Bun present there (alpine + npm).
- Targets: image size reduction >= 300 MB by eliminating ~130 MB Bun binary + installer.
* feat(docker): canonical compose.yaml + smoke-test (P2)
* feat(docker): add canonical compose.yaml for zero-install flow
* ci(docker): smoke-test ccs.kaitran.ca/docker-compose.yaml on release + nightly
* docs(docker): document ccs-net contract for sibling containers
Add "Connect your app to CLIProxy" section to docker/README.md with:
- Public contract table (network=ccs-net, service DNS=ccs, ports 8317/3000)
- Pattern A: same compose file with external network reference
- Pattern B: docker run --network ccs-net
- Troubleshooting subsection (DNS, missing network, conflict, Podman, MTU)
Add one-line link in README.md pointing to the new section.
Add CHANGELOG entry under Unreleased noting the contract as SemVer-major stable.
Add CONTRIBUTING.md note that changing services.ccs or networks.ccs-net requires major bump.
* docs(docker): P3 — hoist two-command quickstart, restructure docker/README, add parity CI
* docs: hoist Docker zero-install quickstart above npm install path
- Create docs/quickstart-snippet.md as canonical source for the
two-command flow (curl + docker compose up -d), wrapped in
<!-- quickstart-snippet-start/end --> markers
- Hoist the snippet into README.md immediately below the deprecation
banner, above all other install paths
- Rename old npm-only "## Quick Start" to "## Install on Host (npm)"
and move it below the Docker quickstart
* docs(docker): restructure README with zero-install first and migration section
- Reorder top-level sections: zero-install (canonical snippet with
markers), choosing an image, power-user ccs docker, prebuilt image,
connect your app to CLIProxy, migration, env vars, troubleshooting
- Add deprecation banner at the top pointing at the migration section
- Add ## Migration from ccs-dashboard:latest section with step-by-step
instructions covering compose down, data preservation, named volume
vs bind-mount path, and compose up with the new image
- Keep P1's Choosing an image table and P5's Connect Your App to
CLIProxy section intact, just repositioned
* test(docs): parity check for quickstart snippet across README files
Assert README.md and docker/README.md both contain the canonical
quickstart block verbatim, anchored by marker comments. Exits non-zero
and prints a diff on any drift.
* ci(docs): wire quickstart-parity test on push and PR
Runs tests/docs/quickstart-parity.sh on self-hosted runner whenever
docs/quickstart-snippet.md, README.md, docker/README.md, or the
test/workflow files themselves change. Fails fast on snippet drift.
* fix(docker): apply red-team findings — drop :full, rc.1 soak, healthcheck, signing
* docs(quickstart): fix raw URL for corporate-proxy fallback (H1)
* feat(docker)!: drop :full image variant — use sibling containers on ccs-net (Q3)
No AI CLIs (claude-code/gemini-cli/grok-cli/opencode) are bundled in the image.
Use sibling containers attached to ccs-net instead.
See docker/README.md#connect-your-app-to-cliproxy.
Also removes bash from apk deps (entrypoint uses #!/bin/sh — L2).
ci(docker): publish only immutable :<ver> tag pre-smoke; promote-mutable-tags
job adds :latest/:MAJOR/:MINOR aliases only after smoke tests pass (H3)
ci(docker): smoke-test-compose-url runs network-contract.sh against the
downloaded /tmp/ccs-compose.yaml instead of re-cloning the repo (H4)
ci(docker): sign published images with cosign keyless OIDC + attach
provenance/SBOM via build-push-action (M8)
test(docker): network-contract.sh now accepts compose-file and image-ref
positional args; replaces python3 healthcheck parser with jq (L4)
* chore(release): cut every main release as rc.N prerelease, manual promote flow (H2)
- .releaserc.cjs: main branch now uses prerelease 'rc' channel — every
semantic-release cut becomes vX.Y.Z-rc.N
- add promote-release.yml: workflow_dispatch flips rc → stable via
'gh release edit --prerelease=false'; triggers docker promote-mutable-tags
- add docs/release-process.md: full soak + promote procedure, rollback steps,
cosign verification command
- releaseNotesGenerator: add revert section, document chore hidden behaviour (L8)
* fix(docker): healthcheck probes both dashboard and cliproxy ports (M1)
compose.yaml healthcheck now checks :3000 and :8317 concurrently with
a 4.5s internal timeout, within Docker's 5s timeout budget.
Also:
- docs(docker): document npm lockfile ephemeral tradeoff above install
layer; note size-budget regression test as the practical safeguard (M3)
- docs(docker): drop :full row from Choosing an image table; add sibling
container note pointing to connect-your-app-to-cliproxy (Q3/docs)
- docs(docker): remove :full docker run block; fix release-tag sentence (Q3)
- docs(docker): fix raw URL in migration section (H1 parity)
- docs(docker): add Volume warning — 'down -v' deletes named volumes (L13)
- docs(docker): update What changes table — remove :full reference (Q3)
- docs(docker): add Image Signatures and SBOM section with cosign verify
and sbom download commands (M8/docs)
- changelog: add Unreleased entries for rc soak, cosign signing, :full
removal with migration guidance
* ci(docker): assert image-size budget per platform; add compose parity + breaking-change guard (M6/L9/L12)
- image-size.sh: add --platform flag; uses 'docker buildx imagetools
inspect' to sum compressed layer sizes from registry manifest for
linux/amd64 and linux/arm64 separately (M6)
- docker-release.yml smoke-test: runs size check for both platforms
- compose-parity.sh: diffs docker/compose.yaml vs
docker/docker-compose.integrated.yml for image name, ports 3000/8317,
volume mounts /root/.ccs and /var/log/ccs, ccs-net definition (L12)
- ci.yml: add compose-parity job wired to cliproxy runner (L12)
- breaking-change-guard.yml: fails PR if compose.yaml changes image name,
network name, or container_name without a feat!/fix! commit (L9)
* chore(ci): fix cosign shell substitution — use tr instead of bash @L expansion (L6/nit)
* test(docker): fix compose-parity port regex for variable-interpolated host ports
* fix: address upstream reviewer findings + failing CI checks (loop 1)
* fix(ci): breaking-change-guard skips missing base files (CI-1)
Guard each git-show call with git cat-file -e existence check before
attempting to read docker/compose.yaml from the base branch. When the
file doesn't exist on the base (new file in this PR), the script would
crash with "fatal: path exists on disk but not in origin/dev". New
files can't cause a contract regression, so we exit early with
breaking=0.
* style: prettier reformat unrelated drift (CI-2)
Two files had minor formatting drift from earlier umbrella PRs:
- src/cliproxy/quota/quota-manager.ts
- src/management/checks/image-analysis-check.ts
No logic changes — formatter-only pass to unblock CI format:check.
* fix(test): update trusted-author gate count to 4 in ci-workflow test (CI-3)
PR added a compose-parity job to ci.yml. That job runs on
self-hosted runners using PR-provided checkout, so it legitimately
requires the trusted-author guard (same as validate, build, test).
The test expected 3 occurrences; the correct count is now 4:
validate (matrix), build, test, compose-parity.
* fix(ci): smoke-test passes compose path + image ref to network-contract (REV-1)
network-contract.sh signature is: <compose-file> [image-ref]
The smoke-test job was calling it as:
bash tests/docker/network-contract.sh "${{ steps.image.outputs.ref }}"
which placed the image ref in the compose-file position ($1), causing
the script to try docker compose -f <image-ref> which fails.
Corrected to:
bash tests/docker/network-contract.sh docker/compose.yaml "${{ steps.image.outputs.ref }}"
Also removes publish-dashboard from smoke-test.needs (REV-2): when
publish-dashboard is SKIPPED on prerelease events, GitHub Actions
propagates the skip to downstream jobs, so smoke-test and
promote-mutable-tags were silently skipped on every rc.N publish.
smoke-test only verifies the integrated image; it has no dependency
on the legacy dashboard image job.
* docs(docker): annotate /root/.ccs path in compose volume (REV-3 clarification)
The reviewer raised a concern that the compose volume mounts /root/.ccs
but the entrypoint might default to /home/node/.ccs. This is a false
positive: the integrated image uses entrypoint-integrated.sh (not
entrypoint.sh), which runs under supervisord with user=root and
explicitly mkdir -p /root/.ccs. HOME is /root inside the container.
The volume mount at /root/.ccs is correct.
Added an inline comment documenting the reasoning so future reviewers
do not confuse entrypoint.sh (legacy dashboard image) with
entrypoint-integrated.sh (integrated image).
* fix(ci): gate docs-parity pull_request job to trusted authors
docs-parity.yml runs on a self-hosted runner and checks out PR code.
The self-hosted-runner-policy test requires any such workflow to include
the trusted-author guard. The workflow was missing the guard, causing
bun test:fast to fail with 1 failure.
Allow push events (no author check needed — push is to own branch)
and trusted-contributor PRs only.
* fix: address reviewer findings round 2 (loop 2)
* fix(docker): parameterize image with CCS_IMAGE env so smoke-test exercises new digest
docker/compose.yaml hardcoded image: ghcr.io/kaitranntt/ccs:latest, causing
network-contract.sh to always use the old published image regardless of what
IMAGE_OVERRIDE was passed. Switching to \${CCS_IMAGE:-ghcr.io/kaitranntt/ccs:latest}
lets CI pass CCS_IMAGE=<digest> so the smoke-test actually exercises the
just-built image. Default end-user behaviour is unchanged.
Resolves REV4 (reviewer loop 2, PR).
* fix(test): image-size.sh fails loudly on manifest inspection failure
The --platform branch previously exited 0 when imagetools inspect returned
empty or zero bytes, silently passing the budget check. A broken image
could reach production undetected if buildx had any inspection issue.
Changed to exit 1 with a clear diagnostic message explaining possible causes
(old buildx, manifest format mismatch, image not yet pushed). No --allow-inspect-failure
escape hatch is provided; CI must fix the root cause.
Resolves REV5 (reviewer loop 2, PR).
* test(docker): cover --platform branch in image-size-logic tests
Prior tests only exercised the local docker image inspect path. The
--platform branch (imagetools inspect) had zero coverage, meaning the
REV5 silent-pass regression would not have been caught by CI.
Added 5 mock-based test cases for the --platform branch:
- pass when platform-scoped compressed size < budget
- fail when platform-scoped compressed size > budget
- fail (exit 1) when imagetools inspect errors (REV5 guard)
- fail (exit 1) when reported size is "0" (REV5 guard)
- fail (exit 1) when size output is empty (REV5 guard)
Mocks follow the existing pattern (override docker in PATH with a temp
wrapper), extended to handle buildx imagetools inspect subcommand.
Resolves REV6 (reviewer loop 2, PR).
* fix(test): compose-parity image_name() handles \${VAR:-default} syntax
After parameterizing compose.yaml's image field with \${CCS_IMAGE:-...},
the image_name() sed pipeline cut at the first colon in the shell variable
syntax (:-) instead of the tag separator, returning '\${CCS_IMAGE' and
failing the kaitranntt/ccs grep.
Added a sed step to unwrap \${VAR:-default} by extracting just the default
value before stripping the tag. Existing plain image: name:tag references
are unaffected.
* fix: tighten parity check + catch service-key rename (loop 3)
* fix(test): tighten compose-parity image-name match to exact expected values (REV7)
Replace loose substring grep (grep -q "ccs") with exact equality checks
against declared EXPECTED_CANONICAL_IMAGE and EXPECTED_INTEGRATED_IMAGE
constants. The integrated compose builds locally as ccs-cliproxy:latest
(not ghcr.io/kaitranntt/ccs), so both expected names are explicitly
documented at the top of the assertion block.
Fixes: a drift in integrated compose to a wrong owner/registry could
slip through the old "grep -q ccs" check; now any name other than the
declared constant is a hard failure.
* feat(ci): breaking-change-guard catches services.ccs rename — public DNS contract (REV8)
Docker's service-name DNS uses the compose service KEY as the hostname.
Sibling containers on ccs-net reach CCS via http://ccs:8317; renaming
services.ccs: to anything else silently breaks that contract even when
image name, network name, and container_name are unchanged.
Add check 4 to the guard:
- Extract top-level service keys from both base and HEAD versions of
docker/compose.yaml using awk (no external YAML parser required).
- Fail if the "ccs" key is absent from HEAD.
- Fail if the sorted set of service keys differs from base.
Wraps inside the existing git cat-file guard so new-file PRs skip it.
* fix(ci): smoke-test failure check + relax service-key guard (REV9, REV10)
REV9 — docker-release.yml smoke-test: add HEALTHY flag to the boot-and-wait
loop. Previously a timeout (status stays 'starting'/'missing' for all 12
iterations) exited the loop silently, letting port probes be the only net.
Now if the loop exits without HEALTHY=1 the step fails immediately with
container logs and state dump. network-contract.sh already has the HEALTHY
pattern — no change needed there.
REV10 — breaking-change-guard.yml: drop the OLD_KEYS != NEW_KEYS comparison.
That check treated ANY change to the service-key set as breaking — including
adding a harmless sidecar. The DNS contract only requires services.ccs to
exist; other services are irrelevant to the 'ccs' hostname on ccs-net. Keep
only the "ccs key must exist" check; remove the unused OLD_KEYS extraction.
* fix(release)!: decouple npm @latest from Docker rc.1 soak (REV11)
Reverts the loop-1 prerelease channel that was publishing npm as
vX.Y.Z-rc.N to the `rc` dist-tag instead of `latest`.
`main` is now a stable semantic-release channel again: every merge
publishes vX.Y.Z immediately to npm @latest. The rc.1 soak window
that guards Docker mutable tags is moved entirely into the Docker
publish workflow:
- `.releaserc.cjs`: remove `prerelease: 'rc'` from productionConfig,
restore `branches: ['main']` (stable). Restore full successComment
and `released` label. Keep loop-1 releaseNotesGenerator additions
(revert section, breaking change comment).
- `docker-release.yml`: every `release: published` event publishes
only the immutable `:<ver>` Docker tag. `promote-mutable-tags` job
now gates exclusively on `workflow_dispatch` with
`promote_to_latest=true` — no longer triggered automatically by
non-prerelease release events.
- `promote-release.yml`: rewritten as a dispatch wrapper that validates
the stable tag exists and is not a prerelease, verifies the immutable
Docker image is in the registry, then dispatches docker-release.yml
with `promote_to_latest=true`. Removes the `gh release edit
--prerelease=false` approach that required the rc soak to be wired
through GitHub release state.
- `docs/release-process.md`: updated to reflect the decoupled model —
npm @latest is immediate; Docker :latest requires manual promote after
soak. Documents the `why` split between npm and Docker soak windows.
* fix(ci): reviewer loop 6 — REV12 compose image parsing, REV13 fork bypass, REV14 :full audit
* fix(ci): breaking-change-guard handles \${VAR:-default} compose image syntax (REV12)
The sed 's/:.*//' pattern truncated at the first colon in the
\${CCS_IMAGE:-ghcr.io/...} expression, yielding "\${CCS_IMAGE" as the
image name instead of the actual registry path. Any change to the default
image namespace was therefore undetectable.
Introduce extract_image_name() that first strips the \${VAR:-default}
wrapper with a sed -E expression, then strips only the trailing :tag
suffix using a pattern that preserves internal colons (e.g. registry:5000/
owner/repo). Applied to both OLD_RAW and NEW_RAW extraction paths.
* fix(ci): breaking-change-guard runs on ubuntu-latest to cover forked PRs (REV13)
The trusted-author gate (COLLABORATOR|MEMBER|OWNER) caused forked-PR
contributors to bypass the breaking-change check entirely. A forked
contributor could rename services.ccs or change the image namespace
without a feat!/fix! marker and the guard would never run.
This workflow is a documented exception to the self-hosted-first policy:
it performs ONLY pure YAML diff parsing (git show / awk / sed). No build,
install, or arbitrary PR-branch scripts are executed. The checkout uses
persist-credentials: false. There is no untrusted code execution, so
ubuntu-latest is safe and necessary for universal fork coverage.
Update self-hosted-runner-policy.test.ts to:
- Introduce GITHUB_HOSTED_RUNNER_EXCEPTIONS registry with required
justification comments for each entry
- Skip exception workflows in the "keeps active workflows on local runners"
and "gates pull-request workflows" assertions
- Add a new "documented exceptions use github-hosted runners" test that
verifies each exception entry actually uses a GitHub-hosted runner
(prevents stale entries accumulating without cleanup)
* docs(ci): clarify rc.1 soak and :full removal as intentional design
Reviewer kept flagging :full omission and stale :latest as bugs. Both are intentional per Q3 maintainer decision and the rc.1 soak design (loop-5). Strengthen inline comments to make this unmistakable to future readers.
* fix: address final-review criticals — healthcheck/digest/contract guards (loop 8)
* fix(docker): healthcheck fails on 4xx not only 5xx (H1)
Change threshold from >= 500 to >= 400 so a misrouted 404 on / is
treated as unhealthy. A container returning 4xx on the root path is
not serving traffic correctly and must not be reported healthy.
* chore(docker): pin eceasy/cli-proxy-api base image by digest (H2)
FROM eceasy/cli-proxy-api:latest@sha256:4fa722b... locks the base image
to a specific manifest-list digest for reproducible builds. The human-
readable :latest tag is preserved for readability. Refresh cadence:
quarterly or on security advisory from upstream. apk packages (nodejs,
npm) remain unpinned — noted as a known limitation until a CI workflow
for apk-digest pinning exists.
* fix(ci): breaking-change-guard catches network rename and hostname override (M1/M2)
Replace the old diff-based ccs-net check (which missed the case where
both old and new lack the name, producing identical empty strings) with
a positive assertion: the canonical compose MUST contain 'name: ccs-net'.
This catches renames, removals, and whitespace variations.
Also add a hostname: override guard (M2): setting services.ccs.hostname
changes the Docker DNS name on ccs-net even when the service key stays
"ccs", silently breaking http://ccs:8317 for sibling containers.
* fix(test): compose-parity image-name regex preserves registry:port (M5)
The old sed 's/:.*//' truncated at the first colon, incorrectly stripping
registry:port prefixes (e.g. registry.local:5000/owner/repo:tag became
registry.local). Replace with sed 's|:[^:/]*$||' which strips only the
trailing :tag suffix, preserving internal colons.
Move tr -d ' ' before the sed -E wrapper-strip so the '^' anchor matches
correctly after whitespace removal. Mirrors extract_image_name() in
breaking-change-guard.yml.
Add 3 image_name() regression tests to image-size-logic.test.sh covering
plain tag, env-var-with-default syntax, and registry:port/owner/repo:tag.
* ci(pr-agent): disable ticket analysis to avoid sub-issue crash
PR-Agent's fetch_sub_issues crashes when the PR body references PR numbers as shortcodes (etc) — GraphQL repository.issue(number) returns null for PR numbers and pr-agent dereferences without a guard (github_provider.py:1048). Disable ticket_analysis_review until upstream fixes the None guard. Other reviewer sections (security, score, tests, etc.) remain enabled.
* ci(pr-agent): re-enable ticket analysis after history+body rewrite
PR shortcodes removed from all commit messages and PR body, so fetch_sub_issues no longer hits the upstream null-deref bug. Ticket compliance review is re-enabled in full.
CCS - Claude Code Switch
The multi-provider profile and runtime manager for Claude Code and compatible CLIs
Run Claude, Codex, Droid-routed profiles, GLM, local models, and Anthropic-compatible APIs without config thrash.
[Docker]
ghcr.io/kaitranntt/ccs-dashboard:latestis deprecated. Useghcr.io/kaitranntt/ccs:latestinstead. See #1251 and docker/README.md for migration details. To wire a sibling container to CLIProxy, see Connect your app to CLIProxy.
Quick Start (Docker)
With Docker installed:
curl -fsSL https://ccs.kaitran.ca/docker-compose.yaml -o docker-compose.yaml
docker compose up -d
Dashboard at http://localhost:3000 · CLIProxy at http://localhost:8317.
Need a corporate-proxy alternative? Download directly:
https://raw.githubusercontent.com/kaitranntt/ccs/main/docker/compose.yaml
Install on Host (npm)
npm install -g @kaitranntt/ccs
ccs config
Then launch whatever runtime fits the task:
ccs
ccs codex
ccs --target droid glm
ccs glm
ccs ollama
Why CCS
CCS gives you one stable command surface while letting you switch between:
- multiple runtimes such as Claude Code, Factory Droid, and Codex CLI
- multiple Claude subscriptions and isolated account contexts
- OAuth providers like Codex, Kiro, Claude, Qwen, Kimi, and more, with legacy Copilot compatibility for existing setups
- API and local-model profiles like GLM, Kimi, OpenRouter, Ollama, llama.cpp, Novita, and Alibaba Coding Plan
The goal is simple: stop rewriting config files, stop breaking active sessions, and move between providers in seconds.
OpenAI-Compatible Routing
CCS can now bridge Claude Code into OpenAI-compatible providers through a local Anthropic-compatible proxy instead of requiring a native Anthropic upstream.
ccs api create --preset hf
ccs hf
Need to manage the proxy manually?
ccs proxy start hf
eval "$(ccs proxy activate)"
The proxy also supports request-time profile:model selectors, scenario-based
model routing through proxy.routing, and explicit activation helpers such as
ccs proxy activate --fish.
Guide: OpenAI-Compatible Provider Routing
Related Project: claude-code-router
claude-code-router is an excellent standalone tool for routing Claude Code requests to OpenAI-compatible providers. CCS's local proxy and SSE transformation work was directly informed by CCR's transformer architecture.
Use CCR when you want a standalone router without CCS profile management.
Use CCS when you want the routing flow integrated with CCS profiles, runtime
bridges, and the existing ccs command surface.
Need the full setup path instead of the short version?
| Need | Start here |
|---|---|
| Install and verify CCS | /getting-started/installation |
| First successful session | /getting-started/first-session |
| Visual walkthrough | /getting-started/product-tour |
| Provider selection | /providers/concepts/overview |
| Full command reference | /reference/cli-commands |
| Troubleshooting | /reference/troubleshooting |
See CCS In Action
Usage Analytics
Track usage, costs, and session patterns across profiles. Deep dive: Dashboard Analytics.
Live Auth And Health Monitoring
See auth state, account health, and provider readiness without dropping into raw config. Deep dive: Live Auth Monitor.
OAuth Provider Control Center
Manage OAuth-backed providers, quota visibility, and proxy-wide routing from one place. CCS now surfaces round-robin vs fill-first natively in both CLI and dashboard flows instead of hiding that choice inside raw upstream controls. The original CLIProxyAPI backend remains the default; the community-maintained CLIProxyAPIPlus fork is opt-in for plus-only providers. When Plus is selected, CCS points the embedded management panel at the maintained CPAMC dashboard fork by default. Deep dive: CLIProxy API.
Managed Tooling And Fallbacks
CCS can provision first-class local tools like WebSearch and image analysis for third-party launches instead of leaving you to wire them by hand. Browser automation now has a first-class setup path as well. Deep dive: WebSearch | Browser Automation.
Docs Matrix
The README stays short on purpose. The docs site owns the detailed guides and reference material.
| If you want to... | Read this |
|---|---|
| Understand what CCS is and how the pieces fit together | Introduction |
| Install CCS cleanly on a new machine | Installation |
| Go from install to a successful first run | Your First CCS Session |
| See the dashboard and workflow surfaces before setup | Product Tour |
| Compare OAuth providers, Claude accounts, and API profiles | Provider Overview |
| Learn the dashboard structure and feature pages | Dashboard Overview |
| Configure profiles, paths, and environment variables | Configuration |
| Understand browser attach vs Codex browser tooling | Browser Automation |
| Keep OpenCode aligned with your live CCS setup | OpenCode Sync Plugin |
| Browse every command and flag | CLI Commands |
| Recover from install, auth, or provider failures | Troubleshooting |
| Understand storage, config, and architecture details | Reference |
Example Workflow
# Design with default Claude
ccs "design the auth flow"
# Implement with a different provider
ccs codex "implement the user service"
# Use a cheaper API profile for routine work
ccs glm "clean up tests and docs"
# Run a local model when you need privacy or offline access
ccs ollama "summarize these logs"
Community Projects
| Project | Author | Description |
|---|---|---|
| opencode-ccs-sync | @JasonLandbridge | Auto-sync CCS providers into OpenCode |
Contribute And Report Safely
- Contributing guide: CONTRIBUTING.md
- Daily local gate:
bun run format && bun run lint:fix && bun run validate(validateis the fast path only) - Before review or merge confidence:
bun run validate:ci-parity - If PR checks stay queued for more than 10 minutes, assume the self-hosted runner is offline and notify a maintainer instead of retrying blindly
- Starter work: good first issue, help wanted
- Questions: open a question issue
- Security reports: SECURITY.md and the private advisory form