Releases: OpenCSGs/csgclaw
v0.3.11
What's Changed
This release expands agent profile management across the web UI and Codex runtime, with stronger avatar and human profile flows, schema-driven runtime configuration, and cleaner participant-based Feishu credential binding.
Highlights:
- Added end-to-end avatar management for agents and humans, including upload/cropping, built-in avatar assignment, and safer save behavior for partial profile edits.
- Introduced human profiles and DM profile actions in the web UI, with avatar updates synced back to IM user records.
- Added agent instructions support for Codex agents and generated managed
AGENTS.mdcontent in local workspaces. - Expanded Codex runtime configuration with schema-driven runtime options, local workspace directory support, host directory picking, and narrower restart behavior.
- Bound Feishu credentials through participants across the CLI, API, provider wiring, and related docs.
Full Changelog
- feat(web): add agent avatar management (#201) (1de03ed)
- fix(web): prevent avatar saves from submitting incomplete profiles (#202) (c1e22bd)
- feat(web): add human profiles and DM profile action (#203) (73fe511)
- feat(agent): support instructions and managed AGENTS.md for codex agents (1ffa681)
- fix(codex): inject env and decode app-server stdout as JSON stream (c26486f)
- feat(web): assign built-in avatars on participant creation (#205) (2cd91c7)
- feat(web): streamline agent profile and creation forms (1e10d48)
- feat(feishu): bind credentials through participants (#204) (135ebcd)
- chore(templates): update embedded manager image tags (929f75b)
- feat(web): show agent skills in profile (7754aa1)
- feat(codex): add schema-driven runtime options for local workspace dir (674f93b)
- refactor(codex): move runtime config handling into the runtime controller (da1792f)
- fix(agent): use field masks for agent patch updates (de185b6)
- feat(api): add host directory picker for agent workspace fields (50da7a5)
- fix(codex): only restart on workspace dir changes (1406561)
- fix(web): inline runtime options into agent profile basics (a684367)
v0.3.10
What's Changed
This release strengthens the platform’s core agent and team model while improving runtime resilience, configuration workflows, and day-to-day web UI behavior.
Highlights:
- Replaced legacy bot identities with a participant-based model across APIs, CLI, runtime wiring, web UI, configs, and docs, with follow-up fixes for legacy participant IDs and admin state migration.
- Added a settings save-and-restart flow so common
config.tomlfields can be edited in the UI, validated before save, and applied with restart guidance. - Improved team and task workflows by refactoring participant mapping, moving task execution logic into
internal/team, and exposing friendly agent display names in task views. - Upgraded Codex and agent runtime behavior by replacing ACP session wiring with the app-server runtime, unifying runtime layout handling, and restoring persisted Codex sessions after server restart.
- Polished operator and user experience with better slash command handling, thread/tool-call rendering fixes, manager setup and upgrade refresh improvements, and safer cleanup/profile save behavior.
Full Changelog
- fix(im): preserve mention context for bot commands (#178) (d8be5e0)
- fix(openclaw): fix plugin permissions and kanban scrolling (#179) (717918d)
- refactor(participant): replace bot identities with participants (#155) (a4cfd35)
- Chore(image): Upgrade PicoClaw image to picoclaw:2026.6.8 (#183) (a748546)
- feat(config): add settings save and restart flow (#182) (caa3994)
- chore(web): tighten frontend config and typings (#181) (b81e060)
- fix(participant): repair legacy agent participant IDs (#184) (da0d4cc)
- ci: allow normal markdown body (#185) (fb7450e)
- fix(participant): migrate admin state and restore upgrade action (#186) (b59d8dc)
- chore: update OpenClaw runtime image tag (#187) (f2ef846)
- docs(license): add logo restriction to LICENSE (fbf79e2)
- feat(web): optimize slash command picker (#190) (83b0666)
- refactor(templates): derive PicoClaw image tags from agent.toml (#188) (8e061bd)
- fix(threads): attach tool calls to the response root (#189) (15ef6b4)
- fix(api): delete unreferenced CSGClaw users on agent cleanup (#191) (cc43ed4)
- Fix(UI): Refresh agent detail after upgrade action (#192) (f216911)
- fix(agent): apply gateway profile on save and fix GLM token limits (#193) (1b08d4b)
- fix(web): show manager setup notice and refresh workspace (#194) (fe514c4)
- feat(ci): add shared Docker embed image pipeline (#195) (9862ca6)
- feat(team): refactor participant mapping and task workflows (#196) (977420c)
- refactor(codex): replace ACP session wiring with app server runtime (0c704e9)
- feat(agent): use embedded templates for image upgrades (#198) (57a5091)
- refactor(agent): unify runtime layout for workspace, skills, and host logs (43c6e73)
- fix(api): separate agent skills browsing from workspace paths (99a4ba5)
- feat(team): expose agent display names in tasks (#200) (44f86e0)
- build(templates): gate image bumps and embed builds (#199) (0db883d)
- fix(codex): restore app-server sessions after server restart (c5c6735)
v0.3.9
What's Changed
This release improves agent and team workflows, expands UI capabilities, and hardens Codex runtime integration and local workspace behavior.
Highlights:
- Added agent avatars end to end, including models, API/service flows, bot creation, and web UI display.
- Improved team task execution and board interactions with clearer planning, task-room projection, parent-child task details, and start controls.
- Added cross-runtime
/newconversation reset support and IM room message clearing. - Improved the web UI with safer agent detail actions, better chat overlay layering, and preserved composer caret behavior during slash tokenization.
- Hardened Codex runtime integration and local execution by stabilizing the provider bridge, isolating runtime shells from host startup environment, and preserving workspace ownership for bind-mounted workspaces.
Full Changelog
- fix(codex): stabilize runtime provider bridge (#171) (5924b4f)
- feat(web): improve agent detail actions (#172) (d81e416)
- feat(agent): add avatar support (#140) (f66d011)
- refactor(ci): embed PicoClaw template dist (#173) (e4f9c9b)
- refactor(agent): resolve workspace roots through registered runtimes (a669992)
- feat(im): add room message clearing (#168) (3a4179c)
- docs: update AGENTS.md (#123) (0a7ca0d)
- refactor(templates): centralize builtin template registry (161f35c)
- fix: preserve workspace ownership and refactor Makefile (#175) (ebc8bc8)
- fix(web): improve chat overlay layering (#176) (84e08d2)
- feat(team): improve task completion and board interactions (#177) (dee5682)
- feat(agent): support /new conversation reset across runtimes (#169) (cb9ba46)
- fix(codex): isolate runtime shells from host startup env (ea8c229)
- fix(web): preserve composer caret after slash tokenization (909d144)
v0.3.8
What's Changed
This release expands OpenClaw integrations, improves slash-command skill workflows, and adds stronger web UI and runtime management support.
Highlights:
- Added Feishu channel support for OpenClaw, aligning it with the existing PicoClaw Feishu integration.
- Expanded slash-command skill workflows with canonical invocation handling, direct-message slash selection, and workspace-backed previews.
- Improved the web UI with workspace file previews, clearer chat timestamps, date dividers, localized labels, and tooltip/sidebar polish.
- Added template-based worker provisioning, dotted hub template refs, local template deletion, and runtime image upgrade flows.
- Fixed agent deletion cleanup by stopping sandboxes before removing agent homes, and updated builtin PicoClaw image refs.
Full Changelog
- fix(agent): stop sandboxes before deleting agent home (#145) (bd87906)
- feat(runtime): support Feishu channel for OpenClaw (#152) (13fa1df)
- feat(skill): add slash picker and workspace preview (#127) (858fad0)
- feat(bot): add template-based worker provisioning (#158) (121a5d6)
- feat(hub): switch template IDs to dotted refs (06151d1)
- feat(slashcommand): support skill invocation commands (#156) (961acdc)
- feat(hub): support deleting local templates (#159) (38c70f7)
- feat(web): improve chat timestamp display (#163) (ebc30b4)
- feat(agent): support runtime image upgrades (#161) (b5151bc)
- feat(web): add workspace file preview dialog (#160) (5c2c1c6)
- fix(web): keep sidebar tooltips above workspace panels (#164) (1367863)
- fix(web): hide tasks sidebar entry (#165) (6c6662d)
- ci(gitlab): build PicoClaw images on release (#166) (9735d37)
- fix(templates): bump builtin picoclaw images to v0.3.7.1 (#170) (39ecf9d)
v0.3.7
What's Changed
This release improves runtime visibility, template-based agent setup, and Web UI polish, while adding deployment controls for upgrade actions.
Highlights:
- Added runtime readiness feedback and a log modal to make agent startup state and logs easier to inspect.
- Applied Hub template environment variables during agent creation so templates can configure agents more completely out of the box.
- Added a
show_upgradeserver setting to hide upgrade actions in managed deployments that should not self-upgrade. - Refined multiple Web UI areas, including message bubbles, workspace sidebar header layout, and the upgrade version menu.
- A podman runtime experiment was introduced and then reverted in the same release range, so it is not part of the effective release outcome.
Full Changelog
- feat: add runtime readiness and log modal (5b51354)
- fix(web): refine message bubble layout (#133) (302ee97)
- fix(web): refine workspace sidebar header layout (#134) (bd8ff62)
- feat(hub): apply template env to agent creation (#95) (c608542)
- feat(sandbox): add podman container runtime support (#136) (bcb8eb1)
- revert(sandbox): add podman container runtime support (#139) (475c77e)
- feat(config): add upgrade visibility setting (#137) (9126a7f)
- fix(web): compact upgrade version menu (#141) (4c5ce14)
v0.3.6
What's Changed
This release expands thread and skill workflows while tightening web UI polish, upgrade handling, and runtime/config behavior.
Highlights:
- Added PicoClaw thread support, including threaded IM sessions, working mentions in threads, and thread replies staying out of the main room timeline.
- Introduced a new
skillCLI command with search, install, version, and registry resolution support for ClawHub-managed skills. - Improved the web UI across several areas: stronger button and form validation styling, agent status indicators in conversations,
cleaner message structure, and smoother sidebar toggle behavior. - Fixed local-build upgrade prompts so development builds no longer show misleading update nags while still reporting the latest release version.
- Reduced runtime/config friction by centralizing default server base URL resolution, improving local IP detection, and fixing manager/thread behavior regressions.
Full Changelog
- refactor(agent): extract local IP detector and avoid tunnel IPv4 selection (e038197)
- fix(model): add per=100 to OpenCSG AIGateway model list (#115) (d3a9821)
- chore(agent): add local IP detection debug logs (f5080f7)
- refactor(config): centralize default server base URL resolution (774e611)
- feat(web): enhance button styles and variants (#114) (b674030)
- fix(upgrade): handle local builds without update prompts (ac6a754)
- chore: add conventional-commit skill (e3e4e81)
- refactor(web): unify UI styling and message structure (#122) (98cb94b)
- ci: update pr-message via Conventional Commits v1.0.0 guidance (#118) (8389ebd)
- feat(threads): support PicoClaw threads (#117) (223874c)
- Fix(manager): Eliminate wrong Need-Recreate flag of Manager (#129) (b3005e9)
- feat(web): add invalid state styling for form controls (#119) (20cd62f)
- Fix(thread): Hide tool calls in threads following the global setting (#130) (2a3f5e8)
- feat(web): show agent status in conversation messages (#107) (47cc0a7)
- fix(web): preserve PicoClaw tool call rendering (#126) (c230e09)
- feat(skill): add skill CLI command (#120) (fcc5f8b)
- feat(web): refine sidebar toggle and slide behavior (#124) (d69a364)
- ci: add GitLab release Docker image build (#111) (37d5972)
v0.3.5
What's Changed
This release improves upgrade behavior, hardens packaging and CI around web assets, and adds runtime, messaging, and web UI refinements.
Highlights:
- Improved upgrade UX by prompting for a manual restart after foreground upgrades and suppressing upgrade checks and UI tips for dev builds.
- Added support for spilling oversized IM session messages into blobs so sessions remain readable under scanner size limits.
- Refined the web UI with stronger button variants and more consistent version display in the sidebar and upgrade modal.
- Hardened release packaging and CI to ensure web assets are built before packaging, including a mirrored Node image for the GitLab web-build job.
- Aligned the PicoClaw runtime workspace layout with the OpenClaw runtime structure for more consistent sandbox and gateway behavior.
Full Changelog
- docs: add weekly release update for v0.3.3-v0.3.4 (04a8e5a)
- fix(release): require built web assets before packaging csgclaw (f187dfa)
- fix(ci): build web assets before GitLab release packaging (17804c9)
- fix(ci): use mirrored Node image for GitLab web-build job (#103) (c793aca)
- refactor(runtime): align PicoClaw workspace layout (#100) (7915100)
- docs: update architecture doc (2f43ca0)
- feat(im): spill oversized session messages to blobs (#104) (0e43613)
- fix(upgrade): skip dev-version upgrade checks and hide UI tips (dbf4eef)
- feat(upgrade): show manual restart prompt after foreground upgrade (6b9f462)
- feat(web): enhance button styles and variants (#108) (189bd5d)
- chore: update builtin picoclaw template image (#110) (f109284)
- feat(web): normalize version display in sidebar and upgrade modal (a8cd0f2)
- update picoclaw template timestamps (#112) (4206214)
- ci: add PR title and body validation workflow (6e1f283)
v0.3.4
What's Changed
This release modernizes the web app foundation, improves sandbox deployment support, adds threaded IM workflows, and introduces user-controlled Codex tool permission approvals.
Highlights:
- Modernized the Web UI with a dedicated Vite/React app, router-first workspace structure, refreshed layout controls, improved Hub browsing, and updated frontend documentation.
- Added release and deployment improvements, including web assets in release packages, deployment-relative asset paths, sandbox-path routing support, and a Dockerfile.
- Moved notification handling to channel bots with relay support, new API routes, supervisor token checks, and notification sidebar UI.
- Added Matrix-shaped thread support across IM storage, API routes, bot compatibility, Codex session isolation, and docs.
- Added Codex tool permission approval flows with structured activity messages, a permission broker, decision API, and Web UI permission cards.
Full Changelog
- refactor(web): modernize frontend structure with Vite/React app (#78) (fdfbd6a)
- docs: add frontend docs and architecture details (#81) (00973ca)
- ci: build web assets for release packages (#82) (35be416)
- feat: refine CSGHub sandbox and add Dockerfile for CSGClaw (#83) (8e567a0)
- feat(web): refine Hub workspace browser styling (#87) (3f52e58)
- fix(web): refine hub template detail layout (#86) (b00018f)
- fix(web): make agent name unchangeable and mark required fields (#84) (a19a2b6)
- refactor(web): modernize workspace routing and structure (#88) (8a6c91a)
- refactor(notification): move notifiers to channel bots (#91) (b58c8a1)
- feat(web): update sidebar brand logo (#92) (5e7a6ee)
- fix(web): support routing under sandbox paths (#94) (ed6e3e0)
- feat(web): refine workspace layout and profile controls (#96) (8b32959)
- feat(im): add threads support (#97) (4dbe75f)
- fix(web): update asset paths to use deployment-relative URLs (#101) (764d023)
- Docs: Add Threads (#102) (0210019)
- feat(codex): support tool permission approvals (#98) (b98a397)
v0.3.3
What's Changed
This release expands template and registry capabilities, improves runtime/API compatibility, and tightens agent and web UI workflows.
Highlights:
- Added Responses API support for LLM providers, with fallback to chat completions where native
/responsessupport is unavailable. - Introduced remote Hub registry support and explicit manager/worker roles in agent templates, improving template discovery and role-aware runtime selection.
- Strengthened agent lifecycle handling by improving manager recreation, making bot deletion more reliable, and surfacing recreate actions more clearly in the web UI.
- Polished operator-facing UX with fixes for notifier state, manager recreate dialogs, worker recreate actions, and Hub dark mode badges.
- Reduced Codex runtime noise by preventing unknown-model metadata warnings for OpenAI-compatible providers.
Full Changelog
- chore: add gh-weekly-release-update skill and release summary doc (23fb477)
- fix(agent): enhance manager recreation logic (#71) (b7d49df)
- fix(web): polish Hub dark mode badges (#70) (da8ef4e)
- chore: update default openclaw image (#74) (23b81e9)
- refactor: require explicit runtime images from bootstrap templates (0166f21)
- chore: update builtin picoclaw template image (26b2560)
- feat(llm): add Responses API support (#72) (6c9edba)
- fix(bot): align web ui list with bots api and harden bot deletion (107b0c1)
- feat(hub): add remote registry support (#77) (eab7dd6)
- fix(web): show recreate button for worker agents (e4ab69f)
- feat(hub): add explicit manager/worker role to agent templates (017131f)
- fix(web): improve manager recreate dialog (854b281)
- fix(codex): avoid unknown-model metadata warnings for Codex runtime (#80) (87acb69)
- fix(web): correct notifier agent UI state from full agent API (#79) (51058d2)
v0.3.2
What's Changed
This release expands runtime capabilities, improves Hub and Web UI workflows, and continues the cleanup of agent, runtime, and API structure.
Highlights:
- Added Notifier runtime support, including webhook delivery, IM fanout, bootstrap/profile integration, and a new
/api/v1/notify/{agent_id}inbound endpoint. - Improved Hub and template workflows by fixing template workspace sync for bot-created agents and restoring/showing Hub template actions in the Web UI.
- Polished the Hub workspace browser and sidebar experience in the Web UI.
- Continued the agent/runtime model cleanup by requiring explicit runtime kinds, consolidating model settings under
agent_profile, and moving runtime setup behind aProvisioner. - Reorganized API routing and channel-scoped bot management, while removing legacy IM routes and updating docs accordingly.
Full Changelog
- chore: add gh-pr-review-comment skill (ef6ed97)
- feat(runtime): add Notifier runtime support (#57) (9fa9866)
- refactor(channel): align multi-channel bot semantics (#56) (2494962)
- feat(web): polish Hub workspace browser UI (#65) (87666d5)
- fix(hub): complete agent template workspace sync (#66) (75573f8)
- refactor(agent): remove runtime kind normalization (b48eab9)
- refactor(agent): enforce explicit runtime kind for workers (28fccc3)
- refactor(agent): consolidate model settings under agent_profile (5149409)
- refactor(notifier): remove legacy runtime option compatibility (beac631)
- refactor(runtime): rename runtime Create to New (7e3d5b2)
- refactor(runtime): introduce Provisioner for runtime setup (d026b52)
- refactor(runtime): move gateway provision inputs into ProvisionRequest (57be92b)
- refactor(api): reorganize HTTP routing and handlers (271c1d9)
- fix: remove unused IM agent-join route and UI entry (cee0698)
- refactor(api): remove legacy /api/v1/im routes (786f61f)
- docs: update API docs (bb41b8c)
- refactor(api): move bot management routes under channel namespaces (ed3a34d)
- fix(api): revert hub template detail route to {id} (0fa89d8)
- fix(web): restore manager recreate action in agent profile (7d2bdea)
- fix(web): show all hub templates in sidebar (f544ceb)