Our review
This skill provides an LLM-driven planner that converts natural-language instructions into structured sub-goal plans for orchestrating visual task execution, with a deterministic keyword-based fallback when no LLM is available.
Strengths
- Flexible decomposition of complex multi-step instructions with dependency handling
- Cross-step references and state management via standardized schemas
- Automatic tool and model discovery
- Robust deterministic fallback when no LLM provider is configured
Limitations
- Requires an LLM provider configured for full intelligent planning capability
- Deterministic fallback may be insufficient for complex or ambiguous instructions
- Risk of over-decomposition for trivial instructions if LLM is not well-guided
Use this skill when you need to execute complex visual generation or editing instructions involving multiple steps and dependencies.
Do not use it for simple single-step instructions that can be handled directly by a single sub-goal without planning.
Security analysis
SafeThe skill is a declarative instruction file that describes a planner component for an orchestration system. It does not contain any destructive commands, exfiltration attempts, or obfuscated payloads. It references existing scripts (propose_plan.py, etc.) that are part of the host system, with no indication of malicious intent or risky operations like arbitrary code execution from untrusted sources.
No concerns found
Examples
Generate an image of a cat sitting on a chair, then animate it to wag its tail.Take the image from my previous turn and add a party hat to the cat.Search for a workflow that creates a 3D render, then run it on the model 'cityscape' and finally evaluate the output quality.name: planning-visual-tasks
description: v1.2 LLM-driven planner. Turns a natural-language instruction into a structured Plan envelope (contracts/agent/plan.schema.json) of typed sub-goals (workflow_search → model_resolve → comfyui_execute → llm_transform → evaluate → mcp_tool → checkpoint → wait_user). Use whenever a user instruction needs routing through the host orchestrator. The legacy v1 keyword router is preserved as a deterministic fallback when no LLM provider is configured.
planning-visual-tasks
v1.2 — the brain of the agent loop. Inputs: a user instruction +
optional task_state + the discovered tool / model / template
inventory. Output: a Plan envelope (matching
contracts/agent/plan.schema.json) that the host
Orchestrator consumes via Executor.run(sub_goal, task_state).
The v1 keyword router (scripts/decompose.py + scripts/emit_dag.py)
is preserved as a deterministic, no-LLM fallback. The host
POST /api/chat/invoke route picks v1.2 when an LLM provider is
configured (host.llm.provider.resolve_credentials() returns a key)
and falls through to the v1 router otherwise. See
host/api/chat_invoke.py::_llm_available and _invoke_stream_v12.
When this Skill applies
- Any natural-language image / video generation, edit, or
composition request that the host has routed through
POST /api/chat/invoke. - Single-goal instructions still go through this Skill — the LLM planner is responsible for the "do not over-decompose" decision and emits a 1-step Plan in that case.
- Multi-step "generate then adjust the light" / "and then make it
dance" instructions become a multi-subgoal DAG with explicit
state://,template://, andmodel://refs across steps. - Cross-turn "and then" continuations consume the prior turn's
TaskState.named_outputsso an earlier image can be referenced by name in the next turn.
Do not invoke when the caller already has a fully-specified
comfyui_execute sub-goal — instantiate the Executor and call
its comfyui_execute handler directly.
Scripts
scripts/propose_plan.py— v1.2 LLM-driven planner. Stdin:{instruction, task_state, available_tools}. Stdout: a Plan envelope. Wrapshost.agent.planner.Planner.propose.scripts/discover_tools.py— stdin{}, stdout{tools, models, templates, mcp_tools}. Lists every executorkindplus stubs from any installed Stream M / W / X surfaces.scripts/validate_plan.py— stdin Plan, stdout{ok, errors}. Checks schema, DAG acyclicity, ref resolution.scripts/legacy/decompose.py— v1 keyword router (no-LLM fallback). Heuristic clause-to-skill mapping. Still callable from the legacy code path for users without an LLM key.scripts/legacy/emit_dag.py— v1 DAG emitter for the legacy router output.scripts/decompose.pyandscripts/emit_dag.py— copies of the legacy scripts kept in place so v1.1 callers (the existinghost.api.chat_invoke.plan_instruction) keep working without modification. Removing these breaks the v1.1 fallback path.
Output contract — v1.2
The Plan envelope shape is canonical
(contracts/agent/plan.schema.json): {plan_id, task_id, user_instruction, execution_order, subgoals[]}. Sub-goal kind
values are limited to: workflow_search, model_resolve, model_download, comfyui_execute, llm_transform, evaluate, mcp_tool, checkpoint, wait_user. Cross-step refs use
scheme://sgN/<name> format only (state, template, model,
artifact schemes); the legacy $sgN.field form is rejected
by the executor's resolve_inputs unless allow_legacy_refs=True
is set explicitly.
The legacy v1 router emits a different shape ({steps, execution_order, created_at} keyed by skill+intent) — that
shape is consumed by host.runner.runner.Runner.run_dag and is
ONLY used on the no-LLM fallback path.
When this Skill applies
- The user issues an instruction that is not a trivial pass-through to one capability Skill (e.g. "generate a cyberpunk portrait, then replace the background with a neon street, then preserve the face across a style transfer"). Any instruction with ≥2 visual goals must go through the planner.
- Single-goal instructions still go through the planner — it is responsible for the "do not over-decompose" decision and emits a 1-step DAG in that case.
- Unusual requests that aren't covered by any bundled capability Skill route
through
searching-comfyui-solutionsfirst; the planner consumes the ranked search results as candidates.
Do not invoke when the caller already has a fully-specified Visual Intent Object for a single Skill — call the capability Skill directly.
Pipeline
- Parse the instruction: extract
goals[], optionalconstraints, optionalinputs.image_refs[]. - If any clause has no matching keyword and the caller supplies
search_candidateson stdin, the planner emits a search-grounded step for it. The planner itself does not shell out tosearch_local.pyorsearch_web.py— the caller (host runner) is responsible for invoking thesearching-comfyui-solutionsSkill scripts and passing the ranked hits in. Each candidate must carryskillandgoalfields declaring the target capability Skill; candidates without them are rejected (the planner fails loudly instead of guessing). Seescripts/decompose.pyVALID_SKILLS/VALID_GOALS. - Run
scripts/decompose.py— heuristics map (goals + constraints + search candidates) → ordered list of{skill, intent, depends_on, evaluators, provenance}. Any clause that cannot be resolved (no keyword match and no usable candidate) causes the script to exit 2 with ano_matcherror; partial plans are rejected. - Run
scripts/emit_dag.py— validates the decomposition, attaches provenance, returns the DAG JSON on stdout. - Caller (host) renders the DAG, gets user approval, then executes step by
step through each capability Skill's
compile.pyand thencomfyui-execution/submit_prompt.py.
Every step carries:
provenance.source—"bundled"|"search:<url>"|"user-supplied"provenance.retrieved_at— ISO 8601 UTC timestamp (ornullfor bundled)provenance.trust_tier— one ofbundled/web_trusted/web_unverifiedper../searching-comfyui-solutions/reference/trust-tiers.md
One-step vs multi-step decision
See reference/strategies.md. Hard rule: if the instruction contains a
single atomic visual action (generate, inpaint, replace bg,
transfer style, preserve face) and no compositional connective
("then", "and after", "while keeping"), emit a 1-step DAG.
Reference files
reference/intent-schema.md— Visual Intent Object fieldsreference/strategies.md— decomposition heuristicsreference/dag-examples.md— four worked examples including the v1 acceptance test
Scripts
scripts/decompose.py— pure Python; stdlib only; keyword + pattern match; no LLM. Input:{instruction, project_state?, search_candidates?}.scripts/emit_dag.py— validates and attaches provenance.
Error handling
Both scripts emit {"error": "...", "code": "..."} on bad input with exit
code 2. Never raise to caller.
References
knowledge/plans/v1/plan_v1.md§1, §7, §8.4knowledge/plans/v1/todo.md§6 (Stage 3)
Next.js App Router Expert
Development
A skill that turns Claude into a Next.js App Router expert.
README Generator
Development
Creates professional and comprehensive README.md files for your projects.
API Documentation Writer
Development
Generates comprehensive API documentation in OpenAPI/Swagger format.