Skip to content

Convert document parsing to an optional skill - #2154

Closed
kshitijk4poor wants to merge 2 commits into
NousResearch:mainfrom
kshitijk4poor:kshitij/document-parse-skill-pr
Closed

Convert document parsing to an optional skill#2154
kshitijk4poor wants to merge 2 commits into
NousResearch:mainfrom
kshitijk4poor:kshitij/document-parse-skill-pr

Conversation

@kshitijk4poor

Copy link
Copy Markdown
Collaborator

Summary

  • remove the document_parse registered tool, config surface, and LiteParse-specific tool wiring
  • keep the useful document parsing guidance as an official optional skill in optional-skills/research/document-parse
  • rewrite the skill around existing Hermes tools: web_extract, read_file, vision_analyze, and execute_code with LiteParse examples

Why

This follows the feedback on #2082: the parsing guidance is useful, but it should live as a skill rather than add schema and config overhead for every user.

Verification

  • python -m pytest tests/tools/test_document_parse_skill.py tests/test_model_tools.py -q
  • python -m pytest tests/ -q
    Full suite still has 4 unrelated failures that also reproduce on origin/main:
    • tests/hermes_cli/test_gateway_service.py::TestEnsureUserSystemdEnv::test_sets_xdg_runtime_dir_when_missing
    • tests/tools/test_web_tools_config.py::TestParallelClientConfig::test_creates_client_with_key
    • tests/tools/test_web_tools_config.py::TestParallelClientConfig::test_no_key_raises_with_helpful_message
    • tests/tools/test_web_tools_config.py::TestParallelClientConfig::test_singleton_returns_same_instance

Supersedes #2082.

@nidhishgajjar

Copy link
Copy Markdown

Orb Code Review (powered by GLM 5.1 on Orb Cloud)

PR #2154 — Convert document parsing to an optional skill

This converts the document parsing guidance into a proper optional skill with SKILL.md, which is a clean organizational improvement.

What I like

  • Well-structured SKILL.md with clear frontmatter metadata, when-to-use guidance, and code patterns.
  • The test validates both frontmatter validity and that the skill references the right tools (execute_code, web_extract, read_file).
  • The trailing whitespace removal in toolsets.py is a nice cleanup.

Observations

  1. SKILL.md references LiteParse but the skill is about using execute_code to call it. This is fine — the skill acts as guidance for when to use execute_code for document parsing rather than registering a new tool. The distinction between this and PR Add document_parse tool with LiteParse fallback #2082 (which adds a dedicated document_parse tool) is clear.

  2. The LiteParse reference URL (https://developers.llamaindex.ai/liteparse/guides/library-usage/) appears in both the skill and the test. If this URL changes, both need updating, but that's a minor concern.

  3. Fallback guidance is thorough — covers all the cases where LiteParse is unavailable.

Verdict

Clean, minimal, well-tested. Good addition to the optional skills catalog.

Summary: Cleanly converts document parsing guidance into an optional skill with proper frontmatter, code examples, and test validation.

Assessment: approve

1 similar comment
@nidhishgajjar

Copy link
Copy Markdown

Orb Code Review (powered by GLM 5.1 on Orb Cloud)

PR #2154 — Convert document parsing to an optional skill

This converts the document parsing guidance into a proper optional skill with SKILL.md, which is a clean organizational improvement.

What I like

  • Well-structured SKILL.md with clear frontmatter metadata, when-to-use guidance, and code patterns.
  • The test validates both frontmatter validity and that the skill references the right tools (execute_code, web_extract, read_file).
  • The trailing whitespace removal in toolsets.py is a nice cleanup.

Observations

  1. SKILL.md references LiteParse but the skill is about using execute_code to call it. This is fine — the skill acts as guidance for when to use execute_code for document parsing rather than registering a new tool. The distinction between this and PR Add document_parse tool with LiteParse fallback #2082 (which adds a dedicated document_parse tool) is clear.

  2. The LiteParse reference URL (https://developers.llamaindex.ai/liteparse/guides/library-usage/) appears in both the skill and the test. If this URL changes, both need updating, but that's a minor concern.

  3. Fallback guidance is thorough — covers all the cases where LiteParse is unavailable.

Verdict

Clean, minimal, well-tested. Good addition to the optional skills catalog.

Summary: Cleanly converts document parsing guidance into an optional skill with proper frontmatter, code examples, and test validation.

Assessment: approve

@alt-glitch alt-glitch added type/refactor Code restructuring, no behavior change P3 Low — cosmetic, nice to have comp/tools Tool registry, model_tools, toolsets tool/skills Skills system (list, view, manage) labels May 3, 2026
@teknium1 teknium1 closed this May 11, 2026
@kshitijk4poor
kshitijk4poor deleted the kshitij/document-parse-skill-pr branch August 5, 2026 07:08
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

comp/tools Tool registry, model_tools, toolsets P3 Low — cosmetic, nice to have tool/skills Skills system (list, view, manage) type/refactor Code restructuring, no behavior change

Projects

None yet

Development

Successfully merging this pull request may close these issues.

4 participants