docs(reference): add Agent Selection Guide for OpenClaw vs Hermes#3659
docs(reference): add Agent Selection Guide for OpenClaw vs Hermes#3659latenighthackathon wants to merge 2 commits into
Conversation
|
Note Reviews pausedIt looks like this branch is under active development. To avoid overwhelming you with review comments due to an influx of new commits, CodeRabbit has automatically paused this review. You can configure this behavior by changing the Use the following commands to manage reviews:
Use the checkboxes below for quick actions:
📝 WalkthroughWalkthroughThis PR adds comprehensive agent selection documentation across NemoClaw's embedded skill resources and public reference documentation. It establishes guidance for choosing between the default OpenClaw agent and the experimental Hermes agent, including onboarding commands, feature comparisons, and sandbox switching instructions. ChangesAgent Selection Documentation
🎯 1 (Trivial) | ⏱️ ~3 minutes Suggested reviewers
Poem
🚥 Pre-merge checks | ✅ 5✅ Passed checks (5 passed)
✏️ Tip: You can configure your own custom pre-merge checks in the settings. ✨ Finishing Touches🧪 Generate unit tests (beta)
Comment |
![coderabbitai[bot]](https://avatars.githubusercontent.com/in/347564?s=60&v=4){kind=small}
There was a problem hiding this comment.
🧹 Nitpick comments (2)
docs/reference/agent-selection.md (2)
85-86: ⚡ Quick winUse a MyST warning admonition for the experimental-status callout.
Line 85–Line 86 is a cautionary note and should be rendered as
:::{warning}instead of bold emphasis in prose.As per coding guidelines: "Use MyST admonitions (:::{tip}, :::{note}, :::{warning}) for callouts, not bold text or blockquotes."
🤖 Prompt for AI Agents
Verify each finding against current code. Fix only still-valid issues, skip the rest with a brief reason, keep changes minimal, and validate. In `@docs/reference/agent-selection.md` around lines 85 - 86, Replace the bolded caution "The Hermes path is currently marked **experimental**. Interfaces, defaults, and supported features may change without notice." with a MyST warning admonition by wrapping that text inside :::{warning} ... ::: so it renders as a warning block (remove the **bold** markup and keep the same plain sentence content within the admonition).
30-37: ⚡ Quick winAdd a one-sentence lead-in under each H2 before the table/list.
Line 30, Line 88, and Line 131 begin sections with content immediately, but this page style requires an introductory sentence under each section heading.
As per coding guidelines: "Sections use H2 and H3, each starting with an introductory sentence."
Also applies to: 88-102, 131-137
🤖 Prompt for AI Agents
Verify each finding against current code. Fix only still-valid issues, skip the rest with a brief reason, keep changes minimal, and validate. In `@docs/reference/agent-selection.md` around lines 30 - 37, Add a one-sentence introductory lead-in immediately below each H2 before the table or list (e.g., under "## TL;DR" and the other H2 headings mentioned in the review) so each section starts with an explanatory sentence; update the markdown around the headings referenced (look for the "## TL;DR" header and the two other H2 headers in this file) to insert a concise sentence that frames the section content, ensuring the style guideline "Sections use H2 and H3, each starting with an introductory sentence" is followed.
🤖 Prompt for all review comments with AI agents
Verify each finding against current code. Fix only still-valid issues, skip the
rest with a brief reason, keep changes minimal, and validate.
Nitpick comments:
In `@docs/reference/agent-selection.md`:
- Around line 85-86: Replace the bolded caution "The Hermes path is currently
marked **experimental**. Interfaces, defaults, and supported features may change
without notice." with a MyST warning admonition by wrapping that text inside
:::{warning} ... ::: so it renders as a warning block (remove the **bold**
markup and keep the same plain sentence content within the admonition).
- Around line 30-37: Add a one-sentence introductory lead-in immediately below
each H2 before the table or list (e.g., under "## TL;DR" and the other H2
headings mentioned in the review) so each section starts with an explanatory
sentence; update the markdown around the headings referenced (look for the "##
TL;DR" header and the two other H2 headers in this file) to insert a concise
sentence that frames the section content, ensuring the style guideline "Sections
use H2 and H3, each starting with an introductory sentence" is followed.
ℹ️ Review info
⚙️ Run configuration
Configuration used: Path: .coderabbit.yaml
Review profile: CHILL
Plan: Enterprise
Run ID: 96ed660f-8002-44ad-9f14-e99a89ca80e2
📒 Files selected for processing (3)
.agents/skills/nemoclaw-user-reference/SKILL.md.agents/skills/nemoclaw-user-reference/references/agent-selection.mddocs/reference/agent-selection.md
latenighthackathon
force-pushed
the
docs/agent-selection-guide
branch
from
051ea2b to
f4349a0
Compare
There was a problem hiding this comment.
🧹 Nitpick comments (2)
docs/reference/agent-selection.md (2)
116-116: ⚡ Quick winRewrite this sentence in active voice.
“Onboarding the other agent under the same name is rejected...” is passive; use an active subject (for example, “NemoClaw rejects onboarding...”).
As per coding guidelines, "Active voice required. Flag passive constructions."
🤖 Prompt for AI Agents
Verify each finding against current code. Fix only still-valid issues, skip the rest with a brief reason, keep changes minimal, and validate. In `@docs/reference/agent-selection.md` at line 116, Replace the passive sentence “Onboarding the other agent under the same name is rejected with a message similar to:” with an active-voice construction; for example, use a subject like the system or agent (e.g., “NemoClaw rejects onboarding another agent that uses the same name, and shows a message similar to:”) so the sentence reads actively and preserves the original meaning.
30-30: ⚡ Quick winAvoid colon-form heading text in section titles.
## TL;DRconflicts with the title formatting rule; rename to a title without a colon-like form (for example,## Summary).As per coding guidelines, "No colons in titles."
🤖 Prompt for AI Agents
Verify each finding against current code. Fix only still-valid issues, skip the rest with a brief reason, keep changes minimal, and validate. In `@docs/reference/agent-selection.md` at line 30, The section heading "## TL;DR" violates the "No colons in titles" rule; rename the heading text (the "## TL;DR" line) to a non-colon-like title such as "## Summary" (or another short, plain-title alternative) so the section title no longer uses colon-like punctuation while preserving the same content.
🤖 Prompt for all review comments with AI agents
Verify each finding against current code. Fix only still-valid issues, skip the
rest with a brief reason, keep changes minimal, and validate.
Nitpick comments:
In `@docs/reference/agent-selection.md`:
- Line 116: Replace the passive sentence “Onboarding the other agent under the
same name is rejected with a message similar to:” with an active-voice
construction; for example, use a subject like the system or agent (e.g.,
“NemoClaw rejects onboarding another agent that uses the same name, and shows a
message similar to:”) so the sentence reads actively and preserves the original
meaning.
- Line 30: The section heading "## TL;DR" violates the "No colons in titles"
rule; rename the heading text (the "## TL;DR" line) to a non-colon-like title
such as "## Summary" (or another short, plain-title alternative) so the section
title no longer uses colon-like punctuation while preserving the same content.
ℹ️ Review info
⚙️ Run configuration
Configuration used: Path: .coderabbit.yaml
Review profile: CHILL
Plan: Enterprise
Run ID: 0bcf4946-96e1-4c89-a530-bccce8380f8e
📒 Files selected for processing (3)
.agents/skills/nemoclaw-user-reference/SKILL.md.agents/skills/nemoclaw-user-reference/references/agent-selection.mddocs/reference/agent-selection.md
✅ Files skipped from review due to trivial changes (1)
- .agents/skills/nemoclaw-user-reference/SKILL.md
latenighthackathon
force-pushed
the
docs/agent-selection-guide
branch
from
f4349a0 to
273aef8
Compare
There was a problem hiding this comment.
🧹 Nitpick comments (2)
docs/reference/agent-selection.md (2)
30-30: ⚡ Quick winRename this heading to remove colon punctuation.
Line 30 uses
TL;DR, which conflicts with the no-colons-in-titles rule for docs headings.As per coding guidelines, "No colons in titles. Flag 'Inference: Cloud and Local' — should be 'Cloud and Local Inference.'"
🤖 Prompt for AI Agents
Verify each finding against current code. Fix only still-valid issues, skip the rest with a brief reason, keep changes minimal, and validate. In `@docs/reference/agent-selection.md` at line 30, Rename the heading "TL;DR" to remove punctuation (e.g., change the "## TL;DR" heading text to "## TLDR" or an equivalent like "## TLDR Summary") in the docs/reference/agent-selection.md file; locate the heading string "TL;DR" and update it so it no longer contains the semicolon to comply with the no-colons-in-titles rule.
27-27: ⚡ Quick winRemove routine bold emphasis (LLM pattern detected).
Lines 27, 42, 47, and 109 use bold for ordinary prose emphasis (
**OpenClaw**,**Hermes**,**CLI**), which the style guide reserves for UI labels, parameter names, and genuine warnings.As per coding guidelines, "Unnecessary bold on routine instructions ... Bold is reserved for UI labels, parameter names, and genuine warnings."
Also applies to: 42-42, 47-47, 109-109
🤖 Prompt for AI Agents
Verify each finding against current code. Fix only still-valid issues, skip the rest with a brief reason, keep changes minimal, and validate. In `@docs/reference/agent-selection.md` at line 27, The document uses bold markup for routine prose (e.g., "**OpenClaw**", "**Hermes**", "**CLI**") which violates the style guide; remove the surrounding double-asterisk emphasis for these occurrences so they read as plain text ("OpenClaw", "Hermes", "CLI") in the content where they are ordinary prose (locations referencing OpenClaw, Hermes, and CLI in agent-selection.md).
🤖 Prompt for all review comments with AI agents
Verify each finding against current code. Fix only still-valid issues, skip the
rest with a brief reason, keep changes minimal, and validate.
Nitpick comments:
In `@docs/reference/agent-selection.md`:
- Line 30: Rename the heading "TL;DR" to remove punctuation (e.g., change the
"## TL;DR" heading text to "## TLDR" or an equivalent like "## TLDR Summary") in
the docs/reference/agent-selection.md file; locate the heading string "TL;DR"
and update it so it no longer contains the semicolon to comply with the
no-colons-in-titles rule.
- Line 27: The document uses bold markup for routine prose (e.g.,
"**OpenClaw**", "**Hermes**", "**CLI**") which violates the style guide; remove
the surrounding double-asterisk emphasis for these occurrences so they read as
plain text ("OpenClaw", "Hermes", "CLI") in the content where they are ordinary
prose (locations referencing OpenClaw, Hermes, and CLI in agent-selection.md).
ℹ️ Review info
⚙️ Run configuration
Configuration used: Path: .coderabbit.yaml
Review profile: CHILL
Plan: Enterprise
Run ID: 984de6dd-b623-4d75-8da2-4a044a9cab5d
📒 Files selected for processing (3)
.agents/skills/nemoclaw-user-reference/SKILL.md.agents/skills/nemoclaw-user-reference/references/agent-selection.mddocs/reference/agent-selection.md
✅ Files skipped from review due to trivial changes (1)
- .agents/skills/nemoclaw-user-reference/SKILL.md
wscurran
added
documentation
|
✨ Thanks for submitting this detailed PR to add an Agent Selection Guide for OpenClaw vs Hermes. This change aims to improve the documentation by providing a clear comparison of the two agents and helping users decide which one to use. |
|
Docs migration heads-up: #3837 has merged the Fern MDX migration cleanup. Legacy documentation source pages under CI now validates the MDX sources directly, so deleted legacy Markdown pages should stay deleted. |
latenighthackathon
force-pushed
the
docs/agent-selection-guide
branch
from
01f5863 to
2285af5
Compare
There was a problem hiding this comment.
Actionable comments posted: 1
🤖 Prompt for all review comments with AI agents
Verify each finding against current code. Fix only still-valid issues, skip the
rest with a brief reason, keep changes minimal, and validate.
Inline comments:
In `@docs/reference/agent-selection.md`:
- Around line 1-18: The page "Agent Selection Guide" is currently authored as
docs/reference/agent-selection.md (legacy Markdown) and uses MyST admonitions;
move this content into a new MDX file (docs/reference/agent-selection.mdx)
preserving frontmatter, then replace any MyST admonition blocks (e.g., :::
{warning} ... :::) with the Fern callout components (<Warning>...</Warning>) and
use <Note> or <Tip> where appropriate; ensure the MDX imports or layout expect
Fern callouts and update any admonition instances referenced (including the ones
around lines noted in the review) to the corresponding <Warning>/<Note>/<Tip>
tags.
🪄 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: Enterprise
Run ID: 2bff693e-e2a1-436d-80c4-e00d62f96e8a
📒 Files selected for processing (3)
.agents/skills/nemoclaw-user-reference/SKILL.md.agents/skills/nemoclaw-user-reference/references/agent-selection.mddocs/reference/agent-selection.md
✅ Files skipped from review due to trivial changes (1)
- .agents/skills/nemoclaw-user-reference/SKILL.md
|
TY @miyoungc ! Updated all of my docs-related PR's including this one to reflect the new changes, cheers! |
latenighthackathon
force-pushed
the
docs/agent-selection-guide
branch
from
c14cec7 to
cfa817b
Compare
latenighthackathon
force-pushed
the
docs/agent-selection-guide
branch
2 times, most recently
from
f0f2292 to
25db9ec
Compare
latenighthackathon
force-pushed
the
docs/agent-selection-guide
branch
2 times, most recently
from
66f9871 to
0a144c1
Compare
Adds docs/reference/agent-selection.mdx comparing the OpenClaw (nemoclaw) and Hermes (nemohermes) agent paths: what each provides, when to choose which, a feature matrix, and how to switch. Regenerates the nemoclaw-user-reference skill (SKILL.md description + agent-selection reference mirror) via scripts/docs-to-skills.py. Rebuilt on current upstream/main as a single signed commit (the prior branch carried an unsigned Update-Branch merge commit and predated the SKILL.md license-field addition). agent-selection.mdx is unchanged from the reviewed revision; the mirror is regenerated with the current tool so it passes the docs-to-skills verification. Signed-off-by: latenighthackathon <latenighthackathon@users.noreply.github.com>
latenighthackathon
force-pushed
the
docs/agent-selection-guide
branch
from
0a144c1 to
c7c89e2
Compare
|
Closing this. NemoClaw's docs now ship separate per-agent OpenClaw and Hermes user guides (#4632) with an agent picker on the home page, which covers the OpenClaw-vs-Hermes selection this guide set out to provide. Cheers! |
3 participants
Summary
Adds
docs/reference/agent-selection.mdx, a single page that answers "should I use OpenClaw or Hermes?" Today that answer is split across the OpenClaw quickstart, the Hermes quickstart, scattered mentions incommands.mdx/inference-options.mdx/ the network policy pages, and a singlenemohermesrow in the CLI selection guide. The new page is the companion reference to the existing CLI Selection Guide: that one explainsnemoclawvsopenshellfor a given operation; this one explains which agent ends up running inside the sandbox.Changes
docs/reference/agent-selection.mdx(new user-facing source page) with a TL;DR table mapping common goals (chat dashboard, OpenAI-compatible API, both side-by-side) to the right entry point.18789dashboard) and Hermes (experimental, OpenAI-compatible API on port8642), with concrete acceptance criteria under "When to choose OpenClaw" / "When to choose Hermes"..mdxfrom the start so it matches the post-docs: remove legacy markdown docs and refresh MDX checks #3837 Fern MDX migration: it uses the<Warning>Fern component for the experimental-status callout, absolute paths for cross-references, and the flattened frontmatter the migrated pages use..agents/skills/nemoclaw-user-reference/SKILL.md(skill index entry) and add.agents/skills/nemoclaw-user-reference/references/agent-selection.md, the auto-generated skill mirror. The mirror is the canonical output ofpython3 scripts/docs-to-skills.py docs/ .agents/skills/ --prefix nemoclaw-user --doc-platform fern-mdx; skill mirrors live at.mdby design (auto-generated from MDX sources) and are not the legacydocs/**/*.mdpages that docs: remove legacy markdown docs and refresh MDX checks #3837 removed.Type of Change
Verification
npx prek run --all-filespassesnpm testpassesnpm run docsbuilds without warnings (doc changes only)Ran:
prek runclean on all changed files; rendered the MDX page and confirmed the tables and<Warning>block render correctly; verified all cross-reference URLs resolve against the post-migration MDX layout; regenerated the skill mirror viadocs-to-skills.py --doc-platform fern-mdxand confirmed no drift from the canonical output. No source code changed, sonpm testis unaffected.Signed-off-by: latenighthackathon latenighthackathon@users.noreply.github.com