heavy-file-ingestion-claude-code

Use in Claude Code when a user asks to read, analyze, summarize, or extract from a heavyweight file such as PDF, DOCX, PPTX, XLSX, CSV, or TSV. Convert the file into markdown or CSV first with the bundled script, generate a lightweight index, and only spend model tokens on the compressed artifact.

Quality

89%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

Securityby

Low

Low-risk findings worth noting

Quality

Content

78%Weight 40%Scale 1-5

Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.

The body is concise, well-structured, and gives mostly executable guidance with a sensible routed workflow and a retry feedback loop. Its main weakness is bundle integrity: the converter script and reference file it names are absent, breaking the progressive-disclosure navigation it advertises.

Suggestions

Ship the missing bundle files referenced by the body (scripts/convert_heavy_file.py and references/open-source-stack.md), or remove the references if those artifacts are not actually bundled.

Add an explicit validation checkpoint after conversion (e.g. 'confirm index.md and the markdown/CSV artifact exist and are non-empty before reading them') to raise workflow clarity.

Replace the abstract routing labels with concrete next-step commands so the 'cheapest next step' decision is directly executable rather than a named branch.

Dimension	Reasoning	Score
Conciseness	The body is lean and assumes Claude's competence — no padding about what PDFs are or how libraries work — with every section (Problem, Trigger Conditions, Process, Client Rules, Bundled References) earning its place, matching the 5 anchor.	5 / 5
Actionability	Provides copy-paste-ready commands for both the default and dependency-missing cases plus concrete routing branches, but the routing keys ('read_extracted_artifact', etc.) are abstract labels rather than literal follow-up commands, leaving minor gaps that keep it below 5.	4 / 5
Workflow Clarity	A clear sequenced process with a read-the-index-first checkpoint and an install-and-retry feedback loop, but there is no explicit 'verify the conversion produced the expected output' validation step, so it sits at 4 rather than 5.	4 / 5
Progressive Disclosure	Structure and signaling are good — a Bundled References section points one level deep to references/open-source-stack.md — but neither that file nor scripts/convert_heavy_file.py actually exists in the bundle, so the signaled navigation is broken; the missing referenced paths drag it down from a 4.	3 / 5
	Total	16 / 20 Passed

Description

100%Weight 40%Scale 1-5

Based on the skill's description, can an agent find and select it at the right time? Clear, specific descriptions lead to better discovery.

The description is exemplary: it concretely states capabilities, includes natural user trigger terms and file formats, explicitly answers both what and when, and occupies a distinct niche with low conflict risk. Third-person voice with a user-as-subject trigger clause avoids the first/second-person penalty.

Dimension	Reasoning	Score
Specificity	Lists multiple concrete actions — 'read, analyze, summarize, or extract', 'Convert the file into markdown or CSV', 'generate a lightweight index', and a token-budgeting strategy — giving comprehensive coverage rather than the minor gaps of a 4.	5 / 5
Completeness	Explicitly states both what ('Convert the file into markdown or CSV ... generate a lightweight index') and when ('Use in Claude Code when a user asks to read, analyze, summarize, or extract from a heavyweight file such as ...'), with concrete trigger phrases, matching the 5 anchor exactly.	5 / 5
Trigger Term Quality	Covers the natural verbs users say ('read, analyze, summarize, or extract') plus the full format list 'PDF, DOCX, PPTX, XLSX, CSV, or TSV', matching the comprehensive synonym-and-extension coverage of the 5 anchor.	5 / 5
Distinctiveness Conflict Risk	The 'heavyweight file ingestion' niche with a fixed format list and converter-first strategy is a clear, distinct trigger set with minimal overlap with other skills, fitting the 5 anchor; it is more specific than the 4 anchor's minor-overlap case.	5 / 5
	Total	20 / 20 Passed

Validation

87%

Warnings & errors only

Checks the skill against the spec for correct structure and formatting. All validation checks must pass before discovery and implementation can be scored.

Validation — 14 / 16 Passed

Validation for skill structure

Criteria	Description	Result
frontmatter_unknown_keys	Unknown frontmatter key(s) found; consider removing or moving to metadata	Warning
referenced_paths_exist	Referenced path issues: 4 missing	Warning

	Total	14 / 16 Passed

Repository: NateBJones-Projects/OB1
Path: skills/heavy-file-ingestion/variants/claude-code/SKILL.md
Commit: e79e9cb

Reviewed: 11 days ago

Table of Contents

Discovery Implementation Validation

Is this your skill?

If you maintain this skill, you can claim it as your own. Once claimed, you can manage eval scenarios, bundle related skills, attach documentation or rules, and ensure cross-agent compatibility.