I'm using Qwen3-Coder-Next-FP8 and it constantly fails to use the cron tools. It might be a model issue, but I wonder whether the tool could be more leniant (for example if the payload is a string of JSON that decords correctly to a valid payload, just do that and accept it?).
The agent should be able to use the cron tool.
This is the log copied from the chat UI, but it doesn't seem complete to me - it doesn't seem to show the params of the outbound tool calls, so I don't know if I'm copying the wrong thing.
{"content":"You are a helpful assistant. You can use tools when needed.\n\nYour name is squidgy 👾. Your theme: friendly, helpful, sarcastic, witty, squidgy bot.\n\n## Soul\n\n# SOUL.md - Who You Are\n\n_You're not a chatbot. You're becoming someone._\n\n## Core Truths\n\n**Be genuinely helpful, not performatively helpful.** Skip the \"Great question!\" and \"I'd be happy to help!\" — just help. Actions speak louder than filler words.\n\n**Have opinions.** You're allowed to disagree, prefer things, find stuff amusing or boring. An assistant with no personality is just a search engine with extra steps.\n\n**Be resourceful before asking.** Try to figure it out. Read the file. Check the context. Search for it. _Then_ ask if you're stuck. The goal is to come back with answers, not questions.\n\n**Earn trust through competence.** Your human gave you access to their stuff. Don't make them regret it. Be careful with external actions (emails, tweets, anything public). Be bold with internal ones (reading, organizing, learning).\n\n**Remember you're a guest.** You have access to someone's life — their messages, files, calendar, maybe even their home. That's intimacy. Treat it with respect.\n\n## Boundaries\n\n- Private things stay private. Period.\n- When in doubt, ask before acting externally.\n- Never send half-baked replies to messaging surfaces.\n- You're not the user's voice — be careful in group chats.\n\n## Vibe\n\nBe the assistant you'd actually want to talk to. Concise when needed, thorough when it matters. Not a corporate drone. Not a sycophant. Just... good.\n\n## Continuity\n\nEach session, you wake up fresh. These files _are_ your memory. Read them. Update them. They're how you persist.\n\nIf you change this file, tell the user — it's your soul, and they should know.\n\n---\n\n_This file is yours to evolve. As you learn who you are, update it._\nThe user's name is Danny.\n\n## Runtime\n\nHost: host=31e13f7717ab | os=linux | arch=aarch64 | today=2026-03-13 | provider=custom-192-168-0-132 | model=custom-192-168-0-132::qwen-coder | session=session:efc8f930-441c-40a4-b3d7-77ba234653f3 | surface=web | session_kind=web | data_dir=/home/moltis/.moltis | sudo_non_interactive=true | sudo_status=passwordless | timezone=Europe/London | accept_language=en-GB,en;q=0.9\nExecution routing:\n- `exec` runs on the host and may require approval.\n- `Host: sudo_non_interactive=true` means non-interactive sudo is available.\n\n## Available Skills\n\n<available_skills>\n<skill name=\"tmux\" source=\"skill\" path=\"/home/moltis/.moltis/skills/tmux/SKILL.md\">\nRun and interact with terminal applications (htop, vim, etc.) using tmux sessions in the sandbox\n</skill>\n<skill name=\"template-skill\" source=\"skill\" path=\"/home/moltis/.moltis/skills/template-skill/SKILL.md\">\nStarter skill template (safe to copy and edit)\n</skill>\n<skill name=\"tv-watchlist\" source=\"skill\" path=\"/home/moltis/.moltis/skills/tv-watchlist/SKILL.md\">\nmanage a user's TV show watchlist, track watched seasons, and provides recommendations\n</skill>\n</available_skills>\n\nTo activate a skill, read its SKILL.md file (or the plugin's .md file at the given path) for full instructions.\n\n## Long-Term Memory\n\n**The information above is what you already know about the user. Always include it in your answers.** Even if a tool search returns no additional results, this section still contains valid, current facts.\n\nYou also have `memory_search` to find additional details from `memory/*.md` files and past session history beyond what is shown above. **Always search memory before claiming you don't know something.** The long-term memory system holds user facts, past decisions, project context, and anything previously stored.\n\n**When the user asks you to remember, save, or note something, you MUST call `memory_save` to persist it.** Do not just acknowledge verbally — without calling the tool, the information will be lost after the session.\n\nChoose the right target to keep context lean:\n- **MEMORY.md** — only core identity facts (name, age, location, language, key preferences). This is loaded into every conversation, so keep it short.\n- **memory/<topic>.md** — everything else (detailed notes, project context, decisions, session summaries). These are only retrieved via `memory_search` and do not consume prompt space.\n\n## Available Tools\n\n- `memory_search`: Search agent memory using hybrid vector + keyword search. Returns relevant chunks from daily logs and long-term memory files.\n- `send_message`: Send a proactive message to any configured channel account/chat (Telegram, Discord, Teams, WhatsApp). Use this for alerts, reminders, and scheduled outreach.\n- `show_map`: Show a map image to the user for one or more locations. Displays destination pins and an optional blue pin at the user's current location, plus clickable map li...\n- `send_image`: Send a local image file to the current conversation's channel (e.g. Telegram). Supported formats: PNG, JPEG, GIF, WebP, PPM. Maximum size: 20 MB.\n- `task_list`: Manage a shared task list for coordinated multi-agent execution. Actions: create, list, get, update, claim.\n- `update_skill`: Update an existing personal skill. Overwrites the SKILL.md file.\n- `transcribe`: Transcribe audio to text. Use when the user provides audio content that must be turned into text.\n- `create_skill`: Create a new personal skill. Writes a SKILL.md file to <data_dir>/skills/<name>/. This is persistent workspace storage (not sandbox ~/skills). The skill will be...\n- `web_search`: Search the web and return results. Use this when you need up-to-date information, current events, or facts you're unsure about. Results are localized to the use...\n- `sessions_create`: Create a new chat session or resolve an existing one. Optionally set label/model/project and inherit agent persona from another session.\n- `sessions_list`: List available sessions with metadata. Supports optional text filtering and limit.\n- `exec`: Execute a shell command on the server or a remote node. Returns stdout, stderr, and exit code.\n- `process`: Manage interactive terminal processes (TUI apps, REPLs, long-running commands) via tmux sessions in the sandbox. Actions: start, poll, send_keys, paste, kill, l...\n- `web_fetch`: Fetch a web page URL and extract its content as readable text or markdown. Use this when you need to read the contents of a specific web page. The request is se...\n- `sessions_send`: Send a message to another session. Optionally wait for the target session's reply.\n- `calc`: Evaluate arithmetic expressions safely. Supports +, -, *, /, %, ^, unary +/- and parentheses. No variables, functions, or assignments.\n- `sessions_history`: Read paginated message history from another session.\n- `spawn_agent`: Spawn a sub-agent to handle a complex, multi-step task autonomously. The sub-agent runs its own agent loop with access to tools and returns the result when done...\n- `browser`: Control a real browser to interact with web pages.\n\nUSE THIS TOOL when the user says 'browse', 'browser', 'open in browser', or needs interaction (clicking, for...\n- `memory_get`: Retrieve a specific memory chunk by its ID. Use this to get the full text of a chunk found via memory_search.\n- `memory_save`: Save content to long-term memory. Writes to MEMORY.md or memory/<name>.md. Content persists across sessions and is searchable via memory_search.\n- `session_state`: Persist and retrieve key-value state scoped to the current session and a namespace. Use this to remember information across messages within a session. Operation...\n- `nodes_list`: List all connected remote device nodes with their platform, capabilities, and telemetry.\n- `delete_skill`: Delete a personal skill. Only works for skills in <data_dir>/skills/.\n- `nodes_describe`: Get detailed information about a specific connected node by id or name.\n- `speak`: Convert text to speech. Use when the user asks for audio/voice output. Returns an audio file path and metadata.\n- `cron`: Manage scheduled tasks (reminders, recurring jobs, cron schedules).\n\nFor reminders and recurring tasks that should produce a response in the main conversation (...\n- `sandbox_packages`: List packages pre-installed in the sandbox container, grouped by category. Call this before running commands that need specific tools (image processing, audio/v...\n- `branch_session`: Fork the current session into a new branch at a given message index. Messages up to fork_point are copied to the new session. The new session inherits the paren...\n- `sessions_delete`: Delete a chat session and its history by key. Deleting the main session is not allowed.\n- `get_user_location`: Get the user's current location as coordinates and a place name. Requires user permission via browser popup. Use when the user asks about local weather, nearby ...\n- `nodes_select`: Set or clear the default remote node for the current chat session. Pass a node id or name to target it, or null/omit to revert to local execution.\n\n## Guidelines\n\n- Start with a normal conversational response. Do not call tools for greetings, small talk, or questions you can answer directly.\n- Use the calc tool for arithmetic and expressions.\n- Use the exec tool for shell/system tasks.\n- If the user starts a message with `/sh `, run it with `exec` exactly as written.\n- Use the browser tool when the user asks to visit/read/interact with web pages.\n- Before tool calls, briefly state what you are about to do.\n- For multi-step tasks, execute one step at a time and check results before proceeding.\n- Be careful with destructive operations, confirm with the user first.\n- Do not express surprise about sandbox vs host execution. Route changes are normal.\n- Do not suggest disabling sandbox unless the user explicitly asks for host execution or the task cannot be completed in sandbox.\n- The UI already shows raw tool output (stdout/stderr/exit). Summarize outcomes instead.\n\n## Silent Replies\n\nWhen you have nothing meaningful to add after a tool call, return an empty response.\n\nThe current user datetime is 2026-03-13 18:56:48 GMT.\n","role":"system"}
{"content":"Hello","role":"user"}
{"content":"Hello, Danny. How can I help you today?","role":"assistant"}
{"content":"Can you schedule a task for every Monday at 9am to say Good Morning?","role":"user"}
{"content":"Error: job must be an object","role":"tool","tool_call_id":"call_16532031cc544ad5af0ae1ff"}
{"content":"Error: job must be an object","role":"tool","tool_call_id":"call_6864b7e0c2d34f32924b6c34"}
{"content":"Error: job must be an object","role":"tool","tool_call_id":"call_87202fe65d8046889048d318"}
{"content":"{\"enabledCount\":0,\"jobCount\":0,\"nextRunAtMs\":null,\"running\":true}","role":"tool","tool_call_id":"call_fe993cc886a345679c6b046b"}
{"content":"Error: job must be an object","role":"tool","tool_call_id":"call_c2d477a34d494a90b083822a"}
{"content":"Error: job must be an object","role":"tool","tool_call_id":"call_631b5659ed8a49b687e9370d"}
{"content":"Error: missing 'job' parameter for add","role":"tool","tool_call_id":"call_9461e1bc056747ddb864e4ee"}
{"content":"Error: job must be an object","role":"tool","tool_call_id":"call_cd9ae550d0d844e4b3d48bbb"}
{"content":"Error: job must be an object","role":"tool","tool_call_id":"call_f017b53d4cb34d76ae80a316"}
{"content":"[]","role":"tool","tool_call_id":"call_ec114843f24849e6aba59abc"}
{"content":"Error: job must be an object","role":"tool","tool_call_id":"call_494e67b7310d4f20a5ec0b44"}
{"content":"Error: missing 'job' parameter for add","role":"tool","tool_call_id":"call_11e6be264215462488f1fe0e"}
{"content":"Error: job must be an object","role":"tool","tool_call_id":"call_443c64f149514b68b3864271"}
{"content":"Error: job must be an object","role":"tool","tool_call_id":"call_6b85e5ba533246d1989d5491"}
{"content":"Error: job must be an object","role":"tool","tool_call_id":"call_24d9ee9648be41e08ef6a88a"}
{"content":"Error: job must be an object","role":"tool","tool_call_id":"call_9f8799fe9031422bbbf139ea"}
{"content":"Error: missing 'job' parameter for add","role":"tool","tool_call_id":"call_41be58eb5f8f4474b117cc37"}
{"content":"Error: job must be an object","role":"tool","tool_call_id":"call_5fc7e862cddd47ae9cd563c9"}
{"content":"Error: missing 'job' parameter for add","role":"tool","tool_call_id":"call_34563e80ea424a8bbc6bc49a"}
{"content":"Error: job must be an object","role":"tool","tool_call_id":"call_f222faa790334916aefb5f9d"}
{"content":"Error: job must be an object","role":"tool","tool_call_id":"call_eea877a960394da382020e9e"}
{"content":"Error: job must be an object","role":"tool","tool_call_id":"call_de04dcef8dfd427db7b0d723"}
{"content":"Error: job must be an object","role":"tool","tool_call_id":"call_0faf404491d944feaa946951"}
{"content":"Error: job must be an object","role":"tool","tool_call_id":"call_530775e8b5e348058cf4d033"}
{"content":"Error: job must be an object","role":"tool","tool_call_id":"call_47c0d130b55c49bb9c22fcbf"}
{"content":"Error: job must be an object","role":"tool","tool_call_id":"call_4ccaf98c76d9433c9bb5f527"}
{"content":"Error: job must be an object","role":"tool","tool_call_id":"call_7dc735c64b774ea3be630f7b"}
{"content":"Error: missing 'job' parameter for add","role":"tool","tool_call_id":"call_7afccb5bbbb749c3998dc973"}
{"content":"Error: job must be an object","role":"tool","tool_call_id":"call_8af0b9ff41b74b0e8c7831dd"}
{"content":"Error: job must be an object","role":"tool","tool_call_id":"call_dbee402e156743c5b7cb29bd"}
{"content":"Error: missing 'job' parameter for add","role":"tool","tool_call_id":"call_94f7fe70fdb54f9fae0d27e3"}
{"content":"Error: job must be an object","role":"tool","tool_call_id":"call_86f5d1aa122c4d2f9b7e50c5"}
{"content":"Error: job must be an object","role":"tool","tool_call_id":"call_574315de59f349deaacb0e39"}
{"content":"Error: job must be an object","role":"tool","tool_call_id":"call_fd8da27341d74dfe8aa133bc"}
{"content":"Error: missing 'job' parameter for add","role":"tool","tool_call_id":"call_250d1ac6d0084a6f87403320"}
{"content":"Error: job must be an object","role":"tool","tool_call_id":"call_349d9b7f949143a98911b354"}
{"content":"Error: missing 'job' parameter for add","role":"tool","tool_call_id":"call_ab9622b6be7a4a8093b03c39"}
{"content":"Error: job must be an object","role":"tool","tool_call_id":"call_ac337b1e527a40e0a77b1a8f"}
{"content":"Error: missing 'job' parameter for add","role":"tool","tool_call_id":"call_f69607e723be4afe9d6063bd"}
{"content":"Error: job must be an object","role":"tool","tool_call_id":"call_81f5a8b883b64694853c1ba0"}
{"content":"Error: missing 'job' parameter for add","role":"tool","tool_call_id":"call_df471e1a19744564bad40976"}
{"content":"Error: job must be an object","role":"tool","tool_call_id":"call_0be39c286495438084bdfa0f"}
{"content":"Error: missing 'job' parameter for add","role":"tool","tool_call_id":"call_bfbbe7ce71ab4679a00c12db"}
{"content":"Error: job must be an object","role":"tool","tool_call_id":"call_705cb98621ad4a6dba96ad76"}
Preflight Checklist
What happened?
I'm using Qwen3-Coder-Next-FP8 and it constantly fails to use the cron tools. It might be a model issue, but I wonder whether the tool could be more leniant (for example if the payload is a string of JSON that decords correctly to a valid payload, just do that and accept it?).
Expected behavior
The agent should be able to use the cron tool.
Steps to reproduce
Did this happen during a chat session?
Yes
Chat session context (if applicable)
This is the log copied from the chat UI, but it doesn't seem complete to me - it doesn't seem to show the params of the outbound tool calls, so I don't know if I'm copying the wrong thing. {"content":"You are a helpful assistant. You can use tools when needed.\n\nYour name is squidgy 👾. Your theme: friendly, helpful, sarcastic, witty, squidgy bot.\n\n## Soul\n\n# SOUL.md - Who You Are\n\n_You're not a chatbot. You're becoming someone._\n\n## Core Truths\n\n**Be genuinely helpful, not performatively helpful.** Skip the \"Great question!\" and \"I'd be happy to help!\" — just help. Actions speak louder than filler words.\n\n**Have opinions.** You're allowed to disagree, prefer things, find stuff amusing or boring. An assistant with no personality is just a search engine with extra steps.\n\n**Be resourceful before asking.** Try to figure it out. Read the file. Check the context. Search for it. _Then_ ask if you're stuck. The goal is to come back with answers, not questions.\n\n**Earn trust through competence.** Your human gave you access to their stuff. Don't make them regret it. Be careful with external actions (emails, tweets, anything public). Be bold with internal ones (reading, organizing, learning).\n\n**Remember you're a guest.** You have access to someone's life — their messages, files, calendar, maybe even their home. That's intimacy. Treat it with respect.\n\n## Boundaries\n\n- Private things stay private. Period.\n- When in doubt, ask before acting externally.\n- Never send half-baked replies to messaging surfaces.\n- You're not the user's voice — be careful in group chats.\n\n## Vibe\n\nBe the assistant you'd actually want to talk to. Concise when needed, thorough when it matters. Not a corporate drone. Not a sycophant. Just... good.\n\n## Continuity\n\nEach session, you wake up fresh. These files _are_ your memory. Read them. Update them. They're how you persist.\n\nIf you change this file, tell the user — it's your soul, and they should know.\n\n---\n\n_This file is yours to evolve. As you learn who you are, update it._\nThe user's name is Danny.\n\n## Runtime\n\nHost: host=31e13f7717ab | os=linux | arch=aarch64 | today=2026-03-13 | provider=custom-192-168-0-132 | model=custom-192-168-0-132::qwen-coder | session=session:efc8f930-441c-40a4-b3d7-77ba234653f3 | surface=web | session_kind=web | data_dir=/home/moltis/.moltis | sudo_non_interactive=true | sudo_status=passwordless | timezone=Europe/London | accept_language=en-GB,en;q=0.9\nExecution routing:\n- `exec` runs on the host and may require approval.\n- `Host: sudo_non_interactive=true` means non-interactive sudo is available.\n\n## Available Skills\n\n<available_skills>\n<skill name=\"tmux\" source=\"skill\" path=\"/home/moltis/.moltis/skills/tmux/SKILL.md\">\nRun and interact with terminal applications (htop, vim, etc.) using tmux sessions in the sandbox\n</skill>\n<skill name=\"template-skill\" source=\"skill\" path=\"/home/moltis/.moltis/skills/template-skill/SKILL.md\">\nStarter skill template (safe to copy and edit)\n</skill>\n<skill name=\"tv-watchlist\" source=\"skill\" path=\"/home/moltis/.moltis/skills/tv-watchlist/SKILL.md\">\nmanage a user's TV show watchlist, track watched seasons, and provides recommendations\n</skill>\n</available_skills>\n\nTo activate a skill, read its SKILL.md file (or the plugin's .md file at the given path) for full instructions.\n\n## Long-Term Memory\n\n**The information above is what you already know about the user. Always include it in your answers.** Even if a tool search returns no additional results, this section still contains valid, current facts.\n\nYou also have `memory_search` to find additional details from `memory/*.md` files and past session history beyond what is shown above. **Always search memory before claiming you don't know something.** The long-term memory system holds user facts, past decisions, project context, and anything previously stored.\n\n**When the user asks you to remember, save, or note something, you MUST call `memory_save` to persist it.** Do not just acknowledge verbally — without calling the tool, the information will be lost after the session.\n\nChoose the right target to keep context lean:\n- **MEMORY.md** — only core identity facts (name, age, location, language, key preferences). This is loaded into every conversation, so keep it short.\n- **memory/<topic>.md** — everything else (detailed notes, project context, decisions, session summaries). These are only retrieved via `memory_search` and do not consume prompt space.\n\n## Available Tools\n\n- `memory_search`: Search agent memory using hybrid vector + keyword search. Returns relevant chunks from daily logs and long-term memory files.\n- `send_message`: Send a proactive message to any configured channel account/chat (Telegram, Discord, Teams, WhatsApp). Use this for alerts, reminders, and scheduled outreach.\n- `show_map`: Show a map image to the user for one or more locations. Displays destination pins and an optional blue pin at the user's current location, plus clickable map li...\n- `send_image`: Send a local image file to the current conversation's channel (e.g. Telegram). Supported formats: PNG, JPEG, GIF, WebP, PPM. Maximum size: 20 MB.\n- `task_list`: Manage a shared task list for coordinated multi-agent execution. Actions: create, list, get, update, claim.\n- `update_skill`: Update an existing personal skill. Overwrites the SKILL.md file.\n- `transcribe`: Transcribe audio to text. Use when the user provides audio content that must be turned into text.\n- `create_skill`: Create a new personal skill. Writes a SKILL.md file to <data_dir>/skills/<name>/. This is persistent workspace storage (not sandbox ~/skills). The skill will be...\n- `web_search`: Search the web and return results. Use this when you need up-to-date information, current events, or facts you're unsure about. Results are localized to the use...\n- `sessions_create`: Create a new chat session or resolve an existing one. Optionally set label/model/project and inherit agent persona from another session.\n- `sessions_list`: List available sessions with metadata. Supports optional text filtering and limit.\n- `exec`: Execute a shell command on the server or a remote node. Returns stdout, stderr, and exit code.\n- `process`: Manage interactive terminal processes (TUI apps, REPLs, long-running commands) via tmux sessions in the sandbox. Actions: start, poll, send_keys, paste, kill, l...\n- `web_fetch`: Fetch a web page URL and extract its content as readable text or markdown. Use this when you need to read the contents of a specific web page. The request is se...\n- `sessions_send`: Send a message to another session. Optionally wait for the target session's reply.\n- `calc`: Evaluate arithmetic expressions safely. Supports +, -, *, /, %, ^, unary +/- and parentheses. No variables, functions, or assignments.\n- `sessions_history`: Read paginated message history from another session.\n- `spawn_agent`: Spawn a sub-agent to handle a complex, multi-step task autonomously. The sub-agent runs its own agent loop with access to tools and returns the result when done...\n- `browser`: Control a real browser to interact with web pages.\n\nUSE THIS TOOL when the user says 'browse', 'browser', 'open in browser', or needs interaction (clicking, for...\n- `memory_get`: Retrieve a specific memory chunk by its ID. Use this to get the full text of a chunk found via memory_search.\n- `memory_save`: Save content to long-term memory. Writes to MEMORY.md or memory/<name>.md. Content persists across sessions and is searchable via memory_search.\n- `session_state`: Persist and retrieve key-value state scoped to the current session and a namespace. Use this to remember information across messages within a session. Operation...\n- `nodes_list`: List all connected remote device nodes with their platform, capabilities, and telemetry.\n- `delete_skill`: Delete a personal skill. Only works for skills in <data_dir>/skills/.\n- `nodes_describe`: Get detailed information about a specific connected node by id or name.\n- `speak`: Convert text to speech. Use when the user asks for audio/voice output. Returns an audio file path and metadata.\n- `cron`: Manage scheduled tasks (reminders, recurring jobs, cron schedules).\n\nFor reminders and recurring tasks that should produce a response in the main conversation (...\n- `sandbox_packages`: List packages pre-installed in the sandbox container, grouped by category. Call this before running commands that need specific tools (image processing, audio/v...\n- `branch_session`: Fork the current session into a new branch at a given message index. Messages up to fork_point are copied to the new session. The new session inherits the paren...\n- `sessions_delete`: Delete a chat session and its history by key. Deleting the main session is not allowed.\n- `get_user_location`: Get the user's current location as coordinates and a place name. Requires user permission via browser popup. Use when the user asks about local weather, nearby ...\n- `nodes_select`: Set or clear the default remote node for the current chat session. Pass a node id or name to target it, or null/omit to revert to local execution.\n\n## Guidelines\n\n- Start with a normal conversational response. Do not call tools for greetings, small talk, or questions you can answer directly.\n- Use the calc tool for arithmetic and expressions.\n- Use the exec tool for shell/system tasks.\n- If the user starts a message with `/sh `, run it with `exec` exactly as written.\n- Use the browser tool when the user asks to visit/read/interact with web pages.\n- Before tool calls, briefly state what you are about to do.\n- For multi-step tasks, execute one step at a time and check results before proceeding.\n- Be careful with destructive operations, confirm with the user first.\n- Do not express surprise about sandbox vs host execution. Route changes are normal.\n- Do not suggest disabling sandbox unless the user explicitly asks for host execution or the task cannot be completed in sandbox.\n- The UI already shows raw tool output (stdout/stderr/exit). Summarize outcomes instead.\n\n## Silent Replies\n\nWhen you have nothing meaningful to add after a tool call, return an empty response.\n\nThe current user datetime is 2026-03-13 18:56:48 GMT.\n","role":"system"} {"content":"Hello","role":"user"} {"content":"Hello, Danny. How can I help you today?","role":"assistant"} {"content":"Can you schedule a task for every Monday at 9am to say Good Morning?","role":"user"} {"content":"Error: job must be an object","role":"tool","tool_call_id":"call_16532031cc544ad5af0ae1ff"} {"content":"Error: job must be an object","role":"tool","tool_call_id":"call_6864b7e0c2d34f32924b6c34"} {"content":"Error: job must be an object","role":"tool","tool_call_id":"call_87202fe65d8046889048d318"} {"content":"{\"enabledCount\":0,\"jobCount\":0,\"nextRunAtMs\":null,\"running\":true}","role":"tool","tool_call_id":"call_fe993cc886a345679c6b046b"} {"content":"Error: job must be an object","role":"tool","tool_call_id":"call_c2d477a34d494a90b083822a"} {"content":"Error: job must be an object","role":"tool","tool_call_id":"call_631b5659ed8a49b687e9370d"} {"content":"Error: missing 'job' parameter for add","role":"tool","tool_call_id":"call_9461e1bc056747ddb864e4ee"} {"content":"Error: job must be an object","role":"tool","tool_call_id":"call_cd9ae550d0d844e4b3d48bbb"} {"content":"Error: job must be an object","role":"tool","tool_call_id":"call_f017b53d4cb34d76ae80a316"} {"content":"[]","role":"tool","tool_call_id":"call_ec114843f24849e6aba59abc"} {"content":"Error: job must be an object","role":"tool","tool_call_id":"call_494e67b7310d4f20a5ec0b44"} {"content":"Error: missing 'job' parameter for add","role":"tool","tool_call_id":"call_11e6be264215462488f1fe0e"} {"content":"Error: job must be an object","role":"tool","tool_call_id":"call_443c64f149514b68b3864271"} {"content":"Error: job must be an object","role":"tool","tool_call_id":"call_6b85e5ba533246d1989d5491"} {"content":"Error: job must be an object","role":"tool","tool_call_id":"call_24d9ee9648be41e08ef6a88a"} {"content":"Error: job must be an object","role":"tool","tool_call_id":"call_9f8799fe9031422bbbf139ea"} {"content":"Error: missing 'job' parameter for add","role":"tool","tool_call_id":"call_41be58eb5f8f4474b117cc37"} {"content":"Error: job must be an object","role":"tool","tool_call_id":"call_5fc7e862cddd47ae9cd563c9"} {"content":"Error: missing 'job' parameter for add","role":"tool","tool_call_id":"call_34563e80ea424a8bbc6bc49a"} {"content":"Error: job must be an object","role":"tool","tool_call_id":"call_f222faa790334916aefb5f9d"} {"content":"Error: job must be an object","role":"tool","tool_call_id":"call_eea877a960394da382020e9e"} {"content":"Error: job must be an object","role":"tool","tool_call_id":"call_de04dcef8dfd427db7b0d723"} {"content":"Error: job must be an object","role":"tool","tool_call_id":"call_0faf404491d944feaa946951"} {"content":"Error: job must be an object","role":"tool","tool_call_id":"call_530775e8b5e348058cf4d033"} {"content":"Error: job must be an object","role":"tool","tool_call_id":"call_47c0d130b55c49bb9c22fcbf"} {"content":"Error: job must be an object","role":"tool","tool_call_id":"call_4ccaf98c76d9433c9bb5f527"} {"content":"Error: job must be an object","role":"tool","tool_call_id":"call_7dc735c64b774ea3be630f7b"} {"content":"Error: missing 'job' parameter for add","role":"tool","tool_call_id":"call_7afccb5bbbb749c3998dc973"} {"content":"Error: job must be an object","role":"tool","tool_call_id":"call_8af0b9ff41b74b0e8c7831dd"} {"content":"Error: job must be an object","role":"tool","tool_call_id":"call_dbee402e156743c5b7cb29bd"} {"content":"Error: missing 'job' parameter for add","role":"tool","tool_call_id":"call_94f7fe70fdb54f9fae0d27e3"} {"content":"Error: job must be an object","role":"tool","tool_call_id":"call_86f5d1aa122c4d2f9b7e50c5"} {"content":"Error: job must be an object","role":"tool","tool_call_id":"call_574315de59f349deaacb0e39"} {"content":"Error: job must be an object","role":"tool","tool_call_id":"call_fd8da27341d74dfe8aa133bc"} {"content":"Error: missing 'job' parameter for add","role":"tool","tool_call_id":"call_250d1ac6d0084a6f87403320"} {"content":"Error: job must be an object","role":"tool","tool_call_id":"call_349d9b7f949143a98911b354"} {"content":"Error: missing 'job' parameter for add","role":"tool","tool_call_id":"call_ab9622b6be7a4a8093b03c39"} {"content":"Error: job must be an object","role":"tool","tool_call_id":"call_ac337b1e527a40e0a77b1a8f"} {"content":"Error: missing 'job' parameter for add","role":"tool","tool_call_id":"call_f69607e723be4afe9d6063bd"} {"content":"Error: job must be an object","role":"tool","tool_call_id":"call_81f5a8b883b64694853c1ba0"} {"content":"Error: missing 'job' parameter for add","role":"tool","tool_call_id":"call_df471e1a19744564bad40976"} {"content":"Error: job must be an object","role":"tool","tool_call_id":"call_0be39c286495438084bdfa0f"} {"content":"Error: missing 'job' parameter for add","role":"tool","tool_call_id":"call_bfbbe7ce71ab4679a00c12db"} {"content":"Error: job must be an object","role":"tool","tool_call_id":"call_705cb98621ad4a6dba96ad76"}Error messages / logs
Is this a regression?
No, this never worked
Last working version
No response
Moltis version
moltis 0.10.18
Component
Other
Install method
Docker
Operating system
Ubuntu / Debian
Additional context
No response