Robotics Idea Discovery Pipeline
Before each reviewer slot, follow Selectable Reviewers:
resolve invocation → .aris/reviewer.json → Claude Code default (Codex MCP)
with reviewer_route.py through the canonical helper resolver, preserving this
slot's tier, fresh contexts, source/image inputs and deterministic gates.
Codex examples/pins below apply only to backend=codex; other routes use their
own first-call/reply contract. Save the route and actual returned identity;
unavailable selections are REVIEW_UNAVAILABLE. Classify acceptance from actual
families/completion/scope. Forward invocation reviewer/model/effort overrides
to reviewer-bearing children. Generation workers are unchanged.
Orchestrate a robotics-specific idea discovery workflow for: $ARGUMENTS
Overview
This skill chains four sub-skills into a single automated pipeline:
/research-lit → /idea-creator (robotics framing) → /novelty-check → /research-review
(survey) (filter + pilot plan) (verify novel) (critical feedback)
But every phase must be grounded in robotics-specific constraints:
- Embodiment: arm, mobile manipulator, drone, humanoid, quadruped, autonomous car, etc.
- Task family: grasping, insertion, locomotion, navigation, manipulation, rearrangement, multi-step planning
- Observation + action interface: RGB/RGB-D/tactile/language; torque/velocity/waypoints/end-effector actions
- Simulator / benchmark availability: simulation-first by default
- Real robot constraints: hardware availability, reset cost, safety, operator time
- Evaluation quality: success rate plus failure cases, safety violations, intervention count, latency, sample efficiency
- Sim2real story: whether the idea can stay in sim, needs offline logs, or truly requires hardware
The goal is not to produce flashy demos. The goal is to produce ideas that are:
- benchmarkable
- falsifiable
- feasible with available robotics infrastructure
- interesting even if the answer is negative
Constants
- MAX_PILOT_IDEAS = 3 — Validate at most 3 top ideas deeply
- PILOT_MODE =
sim-first — Prefer simulation or offline-log pilots before any hardware execution
- REAL_ROBOT_PILOTS =
explicit approval only — Never assume physical robot access or approval
- AUTO_PROCEED = true — If user does not respond at checkpoints, proceed with the best sim-first option
- REVIEWER_MODEL =
gpt-6-astra — External reviewer model via Codex MCP
- TARGET_VENUES = CoRL, RSS, ICRA, IROS, RA-L — Default novelty and reviewer framing
Override inline, e.g. /idea-discovery-robot "bimanual manipulation" — only sim ideas, no real robot or /idea-discovery-robot "drone navigation" — focus on CoRL/RSS, 2 pilot ideas max
Execution Rule
Follow the phases in order. Do not stop after a checkpoint unless:
- the user explicitly says to stop, or
- the user asks to change scope and re-run an earlier phase
If AUTO_PROCEED=true and the user does not respond, continue immediately to the next phase using the strongest sim-first, benchmark-grounded option.
Phase 0: Frame the Robotics Problem
Before generating ideas, extract or infer this Robotics Problem Frame from $ARGUMENTS and local project context:
- Embodiment
- Task family
- Environment type: tabletop, warehouse, home, outdoor, aerial, driving, legged terrain
- Observation modalities
- Action interface / controller abstraction
- Learning regime: RL, imitation, behavior cloning, world model, planning, VLA/VLM, classical robotics, hybrid
- Available assets: simulator, benchmark suite, teleop data, offline logs, existing codebase, real hardware
- Compute budget
- Safety constraints
- Desired contribution type: method, benchmark, diagnosis, systems, sim2real, data curation
If some fields are missing, make explicit assumptions and default to:
- simulation-first
- public benchmark preferred
- no real robot execution
Write this frame into working notes before moving on. Every later decision should reference it.
Phase 1: Robotics Literature Survey
Invoke:
/research-lit "$ARGUMENTS — focus venues: CoRL, RSS, ICRA, IROS, RA-L, TRO, Science Robotics"
Then reorganize the findings using a robotics lens instead of a generic ML lens.
Build a Robotics Landscape Matrix
For each relevant paper, classify:
| Axis |
Examples |
| Embodiment |
single-arm, mobile manipulator, humanoid, drone, quadruped |
| Task |
pick-place, insertion, navigation, locomotion, long-horizon rearrangement |
| Learning setup |
RL, BC, IL, offline RL, world model, planning, diffusion policy |
| Observation |
RGB, RGB-D, proprioception, tactile, language |
| Action abstraction |
torque, joint velocity, end-effector delta pose, waypoint planner |
| Eval regime |
pure sim, sim+real, real-only, offline benchmark |
| Benchmark |
ManiSkill, RLBench, Isaac Lab, Habitat, Meta-World, CALVIN, LIBERO, custom |
| Metrics |
success rate, collision rate, intervention count, path length, latency, energy |
| Main bottleneck |
sample inefficiency, brittleness, reset cost, perception drift, sim2real gap |
Search Priorities
When refining the survey, prioritize:
- recent work from CoRL, RSS, ICRA, IROS, RA-L
- recent arXiv papers from the last 6-12 months
- benchmark papers and follow-up reproductions
- negative-result or diagnosis papers if they reveal system bottlenecks
What to Look For
Do not stop at "who got the best success rate." Explicitly identify:
- recurring failure modes papers do not fix
- benchmarks that are saturated or misleading
- places where embodiment changes invalidate prior conclusions
- methods that only work with privileged observations
- ideas whose reported gains come from reset engineering, reward shaping, or hidden infrastructure
- task families where evaluation quality is weak even if performance numbers look high
Checkpoint: Present the landscape to the user in robotics terms:
🤖 Robotics survey complete. I grouped the field by embodiment, benchmark, action interface, and sim2real setup.
Main gaps:
1. [...]
2. [...]
3. [...]
Should I generate ideas under this framing, or should I narrow to a specific robot / benchmark / modality?
- User approves (or no response + AUTO_PROCEED=true) → proceed to Phase 2 with the best robotics frame.
- User requests changes (e.g. narrower embodiment, different benchmark family, no sim2real, no hardware) → refine the robotics frame, re-run Phase 1, and present again.
Phase 2: Robotics-Specific Idea Generation and Filtering
Generate ideas only after the robotics frame is explicit.
Invoke the existing idea generator, but pass the Robotics Problem Frame and landscape matrix into the prompt so it does not produce generic ML ideas:
/idea-creator "$ARGUMENTS — robotics frame: [paste Robotics Problem Frame] — focus venues: CoRL, RSS, ICRA, IROS, RA-L — benchmark-specific ideas only — sim-first pilots — no real-robot execution without explicit approval — require failure metrics and baseline clarity"
Then rewrite and filter the output using the robotics-specific rules below.
Each candidate idea must include:
- One-sentence summary
- Target embodiment
- Target benchmark / simulator / dataset
- Core bottleneck being addressed
- Minimum sim-first pilot
- Mandatory metrics
- Expected failure mode if the idea does not work
- Whether the idea truly needs real hardware
Good Robotics Idea Patterns
Prefer ideas that:
- expose a real bottleneck in perception-action coupling
- improve robustness under embodiment or environment shift
- reduce operator time, reset cost, or demonstration cost
- strengthen sim2real transfer with measurable mechanisms
- improve recovery, retry behavior, or failure detection
- create a better benchmark, diagnostic, or evaluation protocol
- test an assumption the community repeats but rarely measures
Weak Robotics Idea Patterns
Downrank ideas that are mostly:
- "apply a foundation model / VLM / diffusion model to robot X" with no new bottleneck analysis
- demo-driven but not benchmarkable
- dependent on inaccessible hardware, custom sensors, or massive private datasets
- impossible to evaluate without a months-long infrastructure build
- only interesting if everything works perfectly
Filtering Rules
For each idea, reject or heavily downrank if:
- no concrete simulator or benchmark is available
- no credible baseline exists
- no measurable metric beyond "looks better"
- real robot execution is required but hardware access is unclear
- the setup depends on privileged observations that make the claim weak
- the expected contribution disappears if evaluation is made fair
Checkpoint: Present the ranked robotics ideas before novelty checking:
💡 Robotics ideas generated. Top candidates:
1. [Idea 1] — Embodiment: [...] — Benchmark: [...] — Pilot: sim/offline — Risk: LOW/MEDIUM/HIGH
2. [Idea 2] — Embodiment: [...] — Benchmark: [...] — Pilot: sim/offline — Risk: LOW/MEDIUM/HIGH
3. [Idea 3] — requires hardware / weak benchmark / high risk
Should I carry the top sim-first ideas into novelty checking and external review?
(If no response, I'll continue with the strongest benchmark-grounded ideas.)
- User picks ideas (or no response + AUTO_PROCEED=true) → proceed to Phase 3 with the top sim-first ideas, then continue to Phase 4 and Phase 5.
- User wants different constraints → update the robotics frame and re-run Phase 2.
- User wants narrower scope → go back to Phase 1 with a tighter embodiment / task / benchmark focus.
Phase 3: Feasibility and Pilot Design
For the top ideas, design a minimal validation package.
If the repository already contains a usable simulator, benchmark harness, or offline dataset pipeline, you may validate the top 1-3 ideas there. If not, do not force execution. Produce a concrete pilot plan instead.
By default, pilots should be one of:
- simulation pilot
- offline log / dataset pilot
- analysis-only pilot using existing benchmark outputs
Only propose a real-robot pilot if the user explicitly wants that.
For each surviving idea, specify:
- Embodiment:
- Benchmark / simulator:
- Baselines:
- Pilot type: sim / offline / real
- Compute estimate:
- Human/operator time:
- Success metrics:
- Failure metrics:
- Safety concerns:
- What result would count as positive signal:
- What negative result would still be publishable:
Real Robot Rule
Never auto-proceed to physical robot testing. If an idea needs hardware:
- mark it as
needs physical validation
- design the sim or offline precursor first
- ask for explicit user confirmation before any real-robot step
If no cheap sim/offline pilot exists, keep the idea in the report but label it high execution risk.
After Phase 3, continue to Phase 4 even if you only produced a pilot plan rather than running a pilot. Lack of immediate execution is not a reason to stop the workflow.
Phase 4: Deep Novelty Verification
For each top idea, run:
/novelty-check "[idea description with embodiment + task family + benchmark + sensor stack + controller/policy class + sim2real angle + target venues: CoRL/RSS/ICRA/IROS/RA-L]"
Robotics novelty checks must include:
- embodiment
- task family
- benchmark / simulator
- sensor stack
- controller / policy type
- sim2real or safety angle if relevant
Be especially skeptical of ideas that are just:
- old method + new benchmark
- VLA/VLM + standard manipulation benchmark
- sim2real claim without new transfer mechanism
If the method is not novel but the finding or evaluation protocol is, say that explicitly.
Phase 5: External Robotics Review
Invoke:
/research-review "[top idea with robotics framing, embodiment, benchmark, baselines, pilot plan, evaluation metrics, and sim2real/hardware risks — review as CoRL/RSS/ICRA reviewer]"
Frame the reviewer as a senior CoRL / RSS / ICRA reviewer. Ask them to focus on:
- whether the contribution is really new for robotics, not just ML
- the minimum benchmark package needed for credibility
- whether the sim2real story is justified
- missing baselines or failure analyses
- whether the idea survives realistic infrastructure constraints
Update the report with the reviewer's minimum viable evidence package.
Phase 6: Final Report
Write or update idea-stage/IDEA_REPORT.md with a robotics-specific structure so it stays compatible with downstream workflows.
# Robotics Idea Discovery Report
**Direction**: $ARGUMENTS
**Date**: [today]
**Pipeline**: research-lit → idea-creator (robotics framing) → novelty-check → research-review
## Robotics Problem Frame
- Embodiment:
- Task family:
- Observation / action interface:
- Available assets:
- Constraints:
## Landscape Matrix
[grouped by embodiment, benchmark, and bottleneck]
## Ranked Ideas
### Idea 1: [title] — RECOMMENDED
- Embodiment:
- Benchmark / simulator:
- Bottleneck addressed:
- Pilot type: sim / offline / real
- Positive signal:
- Novelty:
- Reviewer score:
- Hardware risk:
- Next step:
## Eliminated Ideas
- [idea] — killed because benchmark unclear / hardware inaccessible / novelty weak / no fair evaluation
## Evidence Package for the Top Idea
- Required baselines:
- Required metrics:
- Required failure cases:
- Whether real robot evidence is mandatory:
## Next Steps
- [ ] Implement sim-first pilot
- [ ] Run /novelty-check on the final idea wording
- [ ] Only after approval: consider hardware validation
Key Rules
- Simulation first. Hardware is never the default.
- Benchmark specificity is mandatory. No benchmark, no serious idea.
- Evaluation must include failures. Success rate alone is not enough.
- Embodiment matters. Do not assume a result on one robot transfers to another.
- Avoid foundation-model theater. Novel terminology is not novelty.
- Infrastructure realism matters. Operator time, reset burden, and safety count as research constraints.
- If the contribution is mainly diagnostic or evaluative, say so. That can still be publishable.
Composing with Later Work
After this workflow identifies a strong robotics idea:
/idea-discovery-robot "direction" ← you are here
implement sim-first pilot
/run-experiment ← if infrastructure exists
/auto-review-loop "top robotics idea"
If no simulator or benchmark is available yet, stop at the report and ask the user to choose whether to build infrastructure or pivot to a more executable idea.
Output Protocols
Follow these shared protocols for all output files:
1---2name: idea-discovery-robot-33description: "Workflow 1 adaptation for robotics and embodied AI. Orchestrates robotics-aware literature survey, idea generation, novelty check, and critical review to go from a broad robotics direction to benchmark-grounded, simulation-first ideas. Use when user says \"robotics idea discovery\", \"机器人找idea\", \"embodied AI idea\", \"机器人方向探索\", \"sim2real 选题\", or wants ideas for manipulation, locomotion, navigation, drones, humanoids, or general robot learning."4argument-hint: "[robotics-direction]"5allowed-tools: Bash(*), Read, Write, Edit, Grep, Glob, WebSearch, WebFetch, Skill, mcp__codex__codex, mcp__codex__codex-reply6metadata:7 source: https://skillmd.com/skills/wanshuiyin/idea-discovery-robot-38---910# Robotics Idea Discovery Pipeline1112> **Before each reviewer slot**, follow [Selectable Reviewers](../shared-references/reviewer-backends.md):13> resolve invocation → `.aris/reviewer.json` → Claude Code default (Codex MCP)14> with `reviewer_route.py` through the canonical helper resolver, preserving this15> slot's tier, fresh contexts, source/image inputs and deterministic gates.16> Codex examples/pins below apply only to `backend=codex`; other routes use their17> own first-call/reply contract. Save the route and actual returned identity;18> unavailable selections are `REVIEW_UNAVAILABLE`. Classify acceptance from actual19> families/completion/scope. Forward invocation reviewer/model/effort overrides20> to reviewer-bearing children. Generation workers are unchanged.212223Orchestrate a robotics-specific idea discovery workflow for: **$ARGUMENTS**2425## Overview2627This skill chains four sub-skills into a single automated pipeline:2829```30/research-lit → /idea-creator (robotics framing) → /novelty-check → /research-review31 (survey) (filter + pilot plan) (verify novel) (critical feedback)32```3334But every phase must be grounded in robotics-specific constraints:35- **Embodiment**: arm, mobile manipulator, drone, humanoid, quadruped, autonomous car, etc.36- **Task family**: grasping, insertion, locomotion, navigation, manipulation, rearrangement, multi-step planning37- **Observation + action interface**: RGB/RGB-D/tactile/language; torque/velocity/waypoints/end-effector actions38- **Simulator / benchmark availability**: simulation-first by default39- **Real robot constraints**: hardware availability, reset cost, safety, operator time40- **Evaluation quality**: success rate plus failure cases, safety violations, intervention count, latency, sample efficiency41- **Sim2real story**: whether the idea can stay in sim, needs offline logs, or truly requires hardware4243The goal is not to produce flashy demos. The goal is to produce ideas that are:44- benchmarkable45- falsifiable46- feasible with available robotics infrastructure47- interesting even if the answer is negative4849## Constants5051- **MAX_PILOT_IDEAS = 3** — Validate at most 3 top ideas deeply52- **PILOT_MODE = `sim-first`** — Prefer simulation or offline-log pilots before any hardware execution53- **REAL_ROBOT_PILOTS = `explicit approval only`** — Never assume physical robot access or approval54- **AUTO_PROCEED = true** — If user does not respond at checkpoints, proceed with the best sim-first option55- **REVIEWER_MODEL = `gpt-6-astra`** — External reviewer model via Codex MCP56- **TARGET_VENUES = CoRL, RSS, ICRA, IROS, RA-L** — Default novelty and reviewer framing5758> Override inline, e.g. `/idea-discovery-robot "bimanual manipulation" — only sim ideas, no real robot` or `/idea-discovery-robot "drone navigation" — focus on CoRL/RSS, 2 pilot ideas max`5960## Execution Rule6162Follow the phases in order. Do **not** stop after a checkpoint unless:63- the user explicitly says to stop, or64- the user asks to change scope and re-run an earlier phase6566If `AUTO_PROCEED=true` and the user does not respond, continue immediately to the next phase using the strongest **sim-first, benchmark-grounded** option.6768## Phase 0: Frame the Robotics Problem6970Before generating ideas, extract or infer this **Robotics Problem Frame** from `$ARGUMENTS` and local project context:7172- **Embodiment**73- **Task family**74- **Environment type**: tabletop, warehouse, home, outdoor, aerial, driving, legged terrain75- **Observation modalities**76- **Action interface / controller abstraction**77- **Learning regime**: RL, imitation, behavior cloning, world model, planning, VLA/VLM, classical robotics, hybrid78- **Available assets**: simulator, benchmark suite, teleop data, offline logs, existing codebase, real hardware79- **Compute budget**80- **Safety constraints**81- **Desired contribution type**: method, benchmark, diagnosis, systems, sim2real, data curation8283If some fields are missing, make explicit assumptions and default to:84- **simulation-first**85- **public benchmark preferred**86- **no real robot execution**8788Write this frame into working notes before moving on. Every later decision should reference it.8990## Phase 1: Robotics Literature Survey9192Invoke:9394```95/research-lit "$ARGUMENTS — focus venues: CoRL, RSS, ICRA, IROS, RA-L, TRO, Science Robotics"96```9798Then reorganize the findings using a robotics lens instead of a generic ML lens.99100### Build a Robotics Landscape Matrix101102For each relevant paper, classify:103104| Axis | Examples |105|------|----------|106| Embodiment | single-arm, mobile manipulator, humanoid, drone, quadruped |107| Task | pick-place, insertion, navigation, locomotion, long-horizon rearrangement |108| Learning setup | RL, BC, IL, offline RL, world model, planning, diffusion policy |109| Observation | RGB, RGB-D, proprioception, tactile, language |110| Action abstraction | torque, joint velocity, end-effector delta pose, waypoint planner |111| Eval regime | pure sim, sim+real, real-only, offline benchmark |112| Benchmark | ManiSkill, RLBench, Isaac Lab, Habitat, Meta-World, CALVIN, LIBERO, custom |113| Metrics | success rate, collision rate, intervention count, path length, latency, energy |114| Main bottleneck | sample inefficiency, brittleness, reset cost, perception drift, sim2real gap |115116### Search Priorities117118When refining the survey, prioritize:119- recent work from **CoRL, RSS, ICRA, IROS, RA-L**120- recent arXiv papers from the last 6-12 months121- benchmark papers and follow-up reproductions122- negative-result or diagnosis papers if they reveal system bottlenecks123124### What to Look For125126Do not stop at "who got the best success rate." Explicitly identify:127- recurring failure modes papers do not fix128- benchmarks that are saturated or misleading129- places where embodiment changes invalidate prior conclusions130- methods that only work with privileged observations131- ideas whose reported gains come from reset engineering, reward shaping, or hidden infrastructure132- task families where evaluation quality is weak even if performance numbers look high133134**Checkpoint:** Present the landscape to the user in robotics terms:135136```137🤖 Robotics survey complete. I grouped the field by embodiment, benchmark, action interface, and sim2real setup.138139Main gaps:1401. [...]1412. [...]1423. [...]143144Should I generate ideas under this framing, or should I narrow to a specific robot / benchmark / modality?145```146147- **User approves** (or no response + AUTO_PROCEED=true) → proceed to Phase 2 with the best robotics frame.148- **User requests changes** (e.g. narrower embodiment, different benchmark family, no sim2real, no hardware) → refine the robotics frame, re-run Phase 1, and present again.149150## Phase 2: Robotics-Specific Idea Generation and Filtering151152Generate ideas only after the robotics frame is explicit.153154Invoke the existing idea generator, but pass the **Robotics Problem Frame** and landscape matrix into the prompt so it does not produce generic ML ideas:155156```157/idea-creator "$ARGUMENTS — robotics frame: [paste Robotics Problem Frame] — focus venues: CoRL, RSS, ICRA, IROS, RA-L — benchmark-specific ideas only — sim-first pilots — no real-robot execution without explicit approval — require failure metrics and baseline clarity"158```159160Then rewrite and filter the output using the robotics-specific rules below.161162Each candidate idea must include:163- **One-sentence summary**164- **Target embodiment**165- **Target benchmark / simulator / dataset**166- **Core bottleneck being addressed**167- **Minimum sim-first pilot**168- **Mandatory metrics**169- **Expected failure mode if the idea does not work**170- **Whether the idea truly needs real hardware**171172### Good Robotics Idea Patterns173174Prefer ideas that:175- expose a real bottleneck in perception-action coupling176- improve robustness under embodiment or environment shift177- reduce operator time, reset cost, or demonstration cost178- strengthen sim2real transfer with measurable mechanisms179- improve recovery, retry behavior, or failure detection180- create a better benchmark, diagnostic, or evaluation protocol181- test an assumption the community repeats but rarely measures182183### Weak Robotics Idea Patterns184185Downrank ideas that are mostly:186- "apply a foundation model / VLM / diffusion model to robot X" with no new bottleneck analysis187- demo-driven but not benchmarkable188- dependent on inaccessible hardware, custom sensors, or massive private datasets189- impossible to evaluate without a months-long infrastructure build190- only interesting if everything works perfectly191192### Filtering Rules193194For each idea, reject or heavily downrank if:195- no concrete simulator or benchmark is available196- no credible baseline exists197- no measurable metric beyond "looks better"198- real robot execution is required but hardware access is unclear199- the setup depends on privileged observations that make the claim weak200- the expected contribution disappears if evaluation is made fair201202**Checkpoint:** Present the ranked robotics ideas before novelty checking:203204```205💡 Robotics ideas generated. Top candidates:2062071. [Idea 1] — Embodiment: [...] — Benchmark: [...] — Pilot: sim/offline — Risk: LOW/MEDIUM/HIGH2082. [Idea 2] — Embodiment: [...] — Benchmark: [...] — Pilot: sim/offline — Risk: LOW/MEDIUM/HIGH2093. [Idea 3] — requires hardware / weak benchmark / high risk210211Should I carry the top sim-first ideas into novelty checking and external review?212(If no response, I'll continue with the strongest benchmark-grounded ideas.)213```214215- **User picks ideas** (or no response + AUTO_PROCEED=true) → proceed to Phase 3 with the top sim-first ideas, then continue to Phase 4 and Phase 5.216- **User wants different constraints** → update the robotics frame and re-run Phase 2.217- **User wants narrower scope** → go back to Phase 1 with a tighter embodiment / task / benchmark focus.218219## Phase 3: Feasibility and Pilot Design220221For the top ideas, design a **minimal validation package**.222223If the repository already contains a usable simulator, benchmark harness, or offline dataset pipeline, you may validate the top 1-3 ideas there. If not, do **not** force execution. Produce a concrete pilot plan instead.224225By default, pilots should be one of:226- **simulation pilot**227- **offline log / dataset pilot**228- **analysis-only pilot** using existing benchmark outputs229230Only propose a real-robot pilot if the user explicitly wants that.231232For each surviving idea, specify:233234```markdown235- Embodiment:236- Benchmark / simulator:237- Baselines:238- Pilot type: sim / offline / real239- Compute estimate:240- Human/operator time:241- Success metrics:242- Failure metrics:243- Safety concerns:244- What result would count as positive signal:245- What negative result would still be publishable:246```247248### Real Robot Rule249250**Never auto-proceed to physical robot testing.** If an idea needs hardware:251- mark it as `needs physical validation`252- design the sim or offline precursor first253- ask for explicit user confirmation before any real-robot step254255If no cheap sim/offline pilot exists, keep the idea in the report but label it **high execution risk**.256257After Phase 3, continue to Phase 4 even if you only produced a pilot plan rather than running a pilot. Lack of immediate execution is not a reason to stop the workflow.258259## Phase 4: Deep Novelty Verification260261For each top idea, run:262263```264/novelty-check "[idea description with embodiment + task family + benchmark + sensor stack + controller/policy class + sim2real angle + target venues: CoRL/RSS/ICRA/IROS/RA-L]"265```266267Robotics novelty checks must include:268- embodiment269- task family270- benchmark / simulator271- sensor stack272- controller / policy type273- sim2real or safety angle if relevant274275Be especially skeptical of ideas that are just:276- old method + new benchmark277- VLA/VLM + standard manipulation benchmark278- sim2real claim without new transfer mechanism279280If the method is not novel but the **finding** or **evaluation protocol** is, say that explicitly.281282## Phase 5: External Robotics Review283284Invoke:285286```287/research-review "[top idea with robotics framing, embodiment, benchmark, baselines, pilot plan, evaluation metrics, and sim2real/hardware risks — review as CoRL/RSS/ICRA reviewer]"288```289290Frame the reviewer as a senior **CoRL / RSS / ICRA** reviewer. Ask them to focus on:291- whether the contribution is really new for robotics, not just ML292- the minimum benchmark package needed for credibility293- whether the sim2real story is justified294- missing baselines or failure analyses295- whether the idea survives realistic infrastructure constraints296297Update the report with the reviewer's minimum viable evidence package.298299## Phase 6: Final Report300301Write or update `idea-stage/IDEA_REPORT.md` with a robotics-specific structure so it stays compatible with downstream workflows.302303```markdown304# Robotics Idea Discovery Report305306**Direction**: $ARGUMENTS307**Date**: [today]308**Pipeline**: research-lit → idea-creator (robotics framing) → novelty-check → research-review309310## Robotics Problem Frame311- Embodiment:312- Task family:313- Observation / action interface:314- Available assets:315- Constraints:316317## Landscape Matrix318[grouped by embodiment, benchmark, and bottleneck]319320## Ranked Ideas321322### Idea 1: [title] — RECOMMENDED323- Embodiment:324- Benchmark / simulator:325- Bottleneck addressed:326- Pilot type: sim / offline / real327- Positive signal:328- Novelty:329- Reviewer score:330- Hardware risk:331- Next step:332333## Eliminated Ideas334- [idea] — killed because benchmark unclear / hardware inaccessible / novelty weak / no fair evaluation335336## Evidence Package for the Top Idea337- Required baselines:338- Required metrics:339- Required failure cases:340- Whether real robot evidence is mandatory:341342## Next Steps343- [ ] Implement sim-first pilot344- [ ] Run /novelty-check on the final idea wording345- [ ] Only after approval: consider hardware validation346```347348## Key Rules349350- **Simulation first.** Hardware is never the default.351- **Benchmark specificity is mandatory.** No benchmark, no serious idea.352- **Evaluation must include failures.** Success rate alone is not enough.353- **Embodiment matters.** Do not assume a result on one robot transfers to another.354- **Avoid foundation-model theater.** Novel terminology is not novelty.355- **Infrastructure realism matters.** Operator time, reset burden, and safety count as research constraints.356- **If the contribution is mainly diagnostic or evaluative, say so.** That can still be publishable.357358## Composing with Later Work359360After this workflow identifies a strong robotics idea:361362```363/idea-discovery-robot "direction" ← you are here364implement sim-first pilot365/run-experiment ← if infrastructure exists366/auto-review-loop "top robotics idea"367```368369If no simulator or benchmark is available yet, stop at the report and ask the user to choose whether to build infrastructure or pivot to a more executable idea.370371## Output Protocols372373> Follow these shared protocols for all output files:374> - **[Output Versioning Protocol](../shared-references/output-versioning.md)** — write timestamped file first, then copy to fixed name375> - **[Output Manifest Protocol](../shared-references/output-manifest.md)** — log every output to MANIFEST.md376> - **[Output Language Protocol](../shared-references/output-language.md)** — respect the project's language setting