docs: refresh for v0.0.20 and document context window, max tokens, and reasoning env vars#2108
Conversation
…nv vars Adds a "Tune Model Metadata" section to the Switch Inference Models page covering NEMOCLAW_CONTEXT_WINDOW, NEMOCLAW_MAX_TOKENS, and NEMOCLAW_REASONING build-time env vars landed in #1956, plus the --recreate-sandbox flow to apply changes on existing sandboxes. Regenerates the nemoclaw-user-configure-inference skill and syncs the nemoclaw-user-reference troubleshooting reference so autogen output matches the published docs. Signed-off-by: Miyoung Choi <miyoungc@nvidia.com> Made-with: Cursor
📝 WalkthroughWalkthroughDocumentation-only changes across four files adding a new "Tune Model Metadata" section with build-time environment variables ( Changes
Estimated code review effort🎯 2 (Simple) | ⏱️ ~10 minutes Poem
🚥 Pre-merge checks | ✅ 3✅ Passed checks (3 passed)
✏️ Tip: You can configure your own custom pre-merge checks in the settings. ✨ Finishing Touches📝 Generate docstrings
🧪 Generate unit tests (beta)
Comment |
Promotes 0.0.20 to the preferred entry in docs/versions1.json and adds the 0.0.19 entry that was missed during the previous version cut. conf.py regenerates docs/project.json from the preferred entry at build time, so project.json is updated to match. Signed-off-by: Miyoung Choi <miyoungc@nvidia.com> Made-with: Cursor
There was a problem hiding this comment.
Actionable comments posted: 1
🧹 Nitpick comments (3)
.agents/skills/nemoclaw-user-configure-inference/SKILL.md (2)
198-198: Use active voice.The phrase "Invalid values are ignored" uses passive construction.
Consider: "NemoClaw ignores invalid values, and the default bakes into the image."Note: This file is auto-generated from
docs/inference/switch-inference-providers.md.
Fix the source file and regenerate.LLM pattern detected.
🤖 Prompt for AI Agents
Verify each finding against the current code and only fix it if needed. In @.agents/skills/nemoclaw-user-configure-inference/SKILL.md at line 198, Replace the passive sentence "Invalid values are ignored, and the default bakes into the image." with an active-voice version such as "NemoClaw ignores invalid values, and the default bakes into the image." in the source document (the line that currently contains that exact sentence) and then regenerate the auto-generated SKILL.md so the change propagates to .agents/skills/nemoclaw-user-configure-inference/SKILL.md.
192-198: Clarify validation rules and warn about silent failure.The table lists value constraints as "Positive integer (tokens)" but doesn't specify the exact validation pattern.
Based on the implementation,NEMOCLAW_CONTEXT_WINDOWandNEMOCLAW_MAX_TOKENSmust match/^[1-9][0-9]*$/(no leading zeros, no zero value), andNEMOCLAW_REASONINGmust be exactly the stringtrueorfalse.More importantly, when validation fails, the code silently ignores the invalid value without logging a warning or error.
Users who setNEMOCLAW_CONTEXT_WINDOW=abcwill get no indication their value was rejected until they inspect the bakedopenclaw.json.Consider adding a note in the source documentation (
docs/inference/switch-inference-providers.md) that explains:
- Exact validation rules (e.g., "Positive integer with no leading zeros" or "Must be exactly
trueorfalse")- That invalid values fail silently and users should verify the baked config if uncertain
Example:
Invalid values are silently ignored and the image uses the default value instead. To verify that your custom value was accepted, inspect `openclaw.json` inside the sandbox after creation.Note: This file is auto-generated from
docs/inference/switch-inference-providers.md.
Fix the source file and regenerate withpython3 scripts/docs-to-skills.py docs/ .agents/skills/ --prefix nemoclaw-user.🤖 Prompt for AI Agents
Verify each finding against the current code and only fix it if needed. In @.agents/skills/nemoclaw-user-configure-inference/SKILL.md around lines 192 - 198, Update the source docs/inference/switch-inference-providers.md to clarify exact validation rules for NEMOCLAW vars: state that NEMOCLAW_CONTEXT_WINDOW and NEMOCLAW_MAX_TOKENS must match the regex /^[1-9][0-9]*$/ (positive integers, no leading zeros, no zero) and NEMOCLAW_REASONING must be exactly "true" or "false"; add a short note that invalid values are silently ignored (the default is baked into the image) and advise users to verify their accepted values by inspecting openclaw.json inside the sandbox; then regenerate the SKILL.md by running python3 scripts/docs-to-skills.py docs/ .agents/skills/ --prefix nemoclaw-user so the auto-generated file reflects these clarifications.docs/inference/switch-inference-providers.md (1)
145-145: Use active voice.The phrase "Invalid values are ignored" uses passive construction.
Consider: "NemoClaw ignores invalid values, and the default bakes into the image."LLM pattern detected.
🤖 Prompt for AI Agents
Verify each finding against the current code and only fix it if needed. In `@docs/inference/switch-inference-providers.md` at line 145, Replace the passive sentence "Invalid values are ignored, and the default bakes into the image." with an active-voice variant such as "NemoClaw ignores invalid values, and the default bakes into the image." — update the sentence in switch-inference-providers.md by locating the phrase "Invalid values are ignored" and changing it to "NemoClaw ignores invalid values" (leave or slightly adjust the remainder for grammatical flow if needed).
🤖 Prompt for all review comments with AI agents
Verify each finding against the current code and only fix it if needed.
Inline comments:
In `@docs/inference/switch-inference-providers.md`:
- Around line 139-145: Update the docs to distinguish build-time vs runtime
validation: state that during image build (as in onboard logic) invalid env
values are silently ignored and Dockerfile defaults apply, while at sandbox
startup the function apply_model_override() performs runtime validation, writing
detailed errors to stderr and potentially causing startup to fail; explicitly
list the validation rules for NEMOCLAW_CONTEXT_WINDOW and NEMOCLAW_MAX_TOKENS
(must be positive integers) and NEMOCLAW_REASONING (must be exactly "true" or
"false"), and remove the phrasing that implies invalid values are silently
ignored at runtime.
---
Nitpick comments:
In @.agents/skills/nemoclaw-user-configure-inference/SKILL.md:
- Line 198: Replace the passive sentence "Invalid values are ignored, and the
default bakes into the image." with an active-voice version such as "NemoClaw
ignores invalid values, and the default bakes into the image." in the source
document (the line that currently contains that exact sentence) and then
regenerate the auto-generated SKILL.md so the change propagates to
.agents/skills/nemoclaw-user-configure-inference/SKILL.md.
- Around line 192-198: Update the source
docs/inference/switch-inference-providers.md to clarify exact validation rules
for NEMOCLAW vars: state that NEMOCLAW_CONTEXT_WINDOW and NEMOCLAW_MAX_TOKENS
must match the regex /^[1-9][0-9]*$/ (positive integers, no leading zeros, no
zero) and NEMOCLAW_REASONING must be exactly "true" or "false"; add a short note
that invalid values are silently ignored (the default is baked into the image)
and advise users to verify their accepted values by inspecting openclaw.json
inside the sandbox; then regenerate the SKILL.md by running python3
scripts/docs-to-skills.py docs/ .agents/skills/ --prefix nemoclaw-user so the
auto-generated file reflects these clarifications.
In `@docs/inference/switch-inference-providers.md`:
- Line 145: Replace the passive sentence "Invalid values are ignored, and the
default bakes into the image." with an active-voice variant such as "NemoClaw
ignores invalid values, and the default bakes into the image." — update the
sentence in switch-inference-providers.md by locating the phrase "Invalid values
are ignored" and changing it to "NemoClaw ignores invalid values" (leave or
slightly adjust the remainder for grammatical flow if needed).
🪄 Autofix (Beta)
Fix all unresolved CodeRabbit comments on this PR:
- Push a commit to this branch (recommended)
- Create a new PR with the fixes
ℹ️ Review info
⚙️ Run configuration
Configuration used: Path: .coderabbit.yaml
Review profile: CHILL
Plan: Pro Plus
Run ID: a135c013-a162-44a0-bd33-bdef08f2b377
📒 Files selected for processing (3)
.agents/skills/nemoclaw-user-configure-inference/SKILL.md.agents/skills/nemoclaw-user-reference/references/troubleshooting.mddocs/inference/switch-inference-providers.md
| | Variable | Values | Default | | ||
| |---|---|---| | ||
| | `NEMOCLAW_CONTEXT_WINDOW` | Positive integer (tokens) | `131072` | | ||
| | `NEMOCLAW_MAX_TOKENS` | Positive integer (tokens) | `4096` | | ||
| | `NEMOCLAW_REASONING` | `true` or `false` | `false` | | ||
|
|
||
| Invalid values are ignored, and the default bakes into the image. |
There was a problem hiding this comment.
🧩 Analysis chain
🏁 Script executed:
#!/bin/bash
# Verify that the validation code does not log warnings for invalid values
rg -nP -A5 -B5 'NEMOCLAW_CONTEXT_WINDOW|NEMOCLAW_MAX_TOKENS|NEMOCLAW_REASONING' src/lib/onboard.ts | rg -i 'console\.(warn|error|log)'Repository: NVIDIA/NemoClaw
Length of output: 41
🏁 Script executed:
# Search for where these environment variables are used/validated in the codebase
rg -n 'NEMOCLAW_CONTEXT_WINDOW|NEMOCLAW_MAX_TOKENS|NEMOCLAW_REASONING' --type ts --type js src/Repository: NVIDIA/NemoClaw
Length of output: 807
🏁 Script executed:
# Search more broadly including config and setup files
rg -n 'NEMOCLAW_CONTEXT_WINDOW|NEMOCLAW_MAX_TOKENS|NEMOCLAW_REASONING' --type ts --type js --type json --type yamlRepository: NVIDIA/NemoClaw
Length of output: 2211
🏁 Script executed:
# Read the validation implementation in onboard.ts
sed -n '1235,1270p' src/lib/onboard.tsRepository: NVIDIA/NemoClaw
Length of output: 1623
🏁 Script executed:
# Read the test to understand how validation errors are checked
sed -n '361,378p' test/nemoclaw-start.test.tsRepository: NVIDIA/NemoClaw
Length of output: 817
🏁 Script executed:
# Find where apply_model_override function is defined
rg -n 'apply_model_override' --type ts --type sh --type dockerfile src/ test/Repository: NVIDIA/NemoClaw
Length of output: 93
🏁 Script executed:
# Search for where these error messages are generated
rg -n 'must be a positive integer|must be "true" or "false"' src/Repository: NVIDIA/NemoClaw
Length of output: 41
🏁 Script executed:
# Search for apply_model_override without specifying dockerfile type
rg -n 'apply_model_override' --type ts --type sh src/ test/Repository: NVIDIA/NemoClaw
Length of output: 2793
🏁 Script executed:
# Look at more context in the test file to understand what 'src' is
sed -n '335,365p' test/nemoclaw-start.test.tsRepository: NVIDIA/NemoClaw
Length of output: 1247
🏁 Script executed:
# Look at the start of the test file to understand what 'src' is
head -50 test/nemoclaw-start.test.tsRepository: NVIDIA/NemoClaw
Length of output: 2310
🏁 Script executed:
# Also look for where the apply_model_override function is actually generated
rg -n 'apply_model_override.*{' src/ --type ts -A 10 | head -40Repository: NVIDIA/NemoClaw
Length of output: 168
🏁 Script executed:
# Search for where apply_model_override is generated in the codebase
rg -n 'apply_model_override' src/ --type ts -B 2 -A 2Repository: NVIDIA/NemoClaw
Length of output: 41
🏁 Script executed:
# Let's look at the actual script that the test is checking
head -100 scripts/nemoclaw-start.sh | tail -50Repository: NVIDIA/NemoClaw
Length of output: 2239
🏁 Script executed:
# Find where the error messages are generated in the TypeScript code
rg -n '".*must be.*"' src/lib/ --type tsRepository: NVIDIA/NemoClaw
Length of output: 866
🏁 Script executed:
# Look for where the shell script is generated/written
rg -n 'apply_model_override' src/lib/ --type ts -B 5 -A 5Repository: NVIDIA/NemoClaw
Length of output: 41
🏁 Script executed:
# Search for the error messages in shell script templates
rg -n 'must be a positive integer' --type sh --type tsRepository: NVIDIA/NemoClaw
Length of output: 758
🏁 Script executed:
# Look for where apply_model_override function might be generated or templated
rg -n 'apply_model_override' src/lib/onboard.ts -B 5 -A 10Repository: NVIDIA/NemoClaw
Length of output: 41
🏁 Script executed:
# Check if nemoclaw-start.sh is a static file or generated
ls -la scripts/nemoclaw-start.sh
file scripts/nemoclaw-start.sh
head -20 scripts/nemoclaw-start.shRepository: NVIDIA/NemoClaw
Length of output: 1429
🏁 Script executed:
# Look at the actual apply_model_override function in the shell script
sed -n '250,310p' scripts/nemoclaw-start.shRepository: NVIDIA/NemoClaw
Length of output: 2583
🏁 Script executed:
# Check for NEMOCLAW_REASONING validation in the shell script
sed -n '280,300p' scripts/nemoclaw-start.shRepository: NVIDIA/NemoClaw
Length of output: 937
Consider clarifying validation behavior — values are not silently ignored at runtime.
The current text states "Invalid values are ignored, and the default bakes into the image," which accurately describes build-time behavior (in src/lib/onboard.ts). However, this may mislead readers about runtime validation: at sandbox startup, the apply_model_override() function validates these environment variables and logs detailed errors to stderr if they are invalid, rather than silently ignoring them.
For clarity, update the explanation to distinguish build-time and runtime:
- Build-time (image creation): Invalid values are silently ignored; defaults from the Dockerfile apply.
- Runtime (sandbox startup): Invalid values trigger validation errors logged to stderr; startup may fail depending on how the error is handled.
You may also clarify the exact validation rules: NEMOCLAW_CONTEXT_WINDOW and NEMOCLAW_MAX_TOKENS must be positive integers; NEMOCLAW_REASONING must be exactly "true" or "false".
🤖 Prompt for AI Agents
Verify each finding against the current code and only fix it if needed.
In `@docs/inference/switch-inference-providers.md` around lines 139 - 145, Update
the docs to distinguish build-time vs runtime validation: state that during
image build (as in onboard logic) invalid env values are silently ignored and
Dockerfile defaults apply, while at sandbox startup the function
apply_model_override() performs runtime validation, writing detailed errors to
stderr and potentially causing startup to fail; explicitly list the validation
rules for NEMOCLAW_CONTEXT_WINDOW and NEMOCLAW_MAX_TOKENS (must be positive
integers) and NEMOCLAW_REASONING (must be exactly "true" or "false"), and remove
the phrasing that implies invalid values are silently ignored at runtime.
ericksoa
left a comment
There was a problem hiding this comment.
Docs-only change, clean diff, CI green. LGTM.
Summary
Documents the
NEMOCLAW_CONTEXT_WINDOW,NEMOCLAW_MAX_TOKENS, andNEMOCLAW_REASONINGbuild-time environment variables that landed in #1956. These variables now propagate through the Dockerfile ARG chain intoopenclaw.json, but were not mentioned anywhere in the published docs.Changes
docs/inference/switch-inference-providers.mdwith a variable reference table, defaults, value constraints, an example, and thenemoclaw onboard --resume --recreate-sandboxflow for applying changes to existing sandboxes..agents/skills/nemoclaw-user-configure-inference/SKILL.mdfrom the updated source..agents/skills/nemoclaw-user-reference/references/troubleshooting.mdwith the currentdocs/reference/troubleshooting.md(brings the autogen reference back in line with theopenclaw updateentry added in docs: catch up documentation for v0.0.18 changes #2033).Type of Change
Verification
npx prek run --all-filespassesnpm testpassesmake docsbuilds without warnings (doc changes only)AI Disclosure
Signed-off-by: Miyoung Choi miyoungc@nvidia.com
Made with Cursor
Summary by CodeRabbit