// Unpack a ZIP archive, inventory its files, run the corresponding child grounding skill for each supported child file, and then write a real archive-level grounded.md.
Convert a raw document into a structured grounding note for downstream research and summarization.
Run focused literature and web research from a grounded note. Use when a grounded note already exists and you want targeted research results, opened-link evidence, deeper per-paper analysis materials, optional downloaded literature, and a two-stage literature output (`lit_initial.md` then refined `lit.md`).
Review a research report draft with a structured scoring rubric, run a bounded repair loop when needed, and produce the final deliverable report.
Create a rich, evidence-preserving research report draft from a grounded note and its follow-up literature result. This is the main report-writing stage of the middle pipeline, not a compression memo.
Use the input path to select the correct downstream grounding pipeline and continue execution until the selected grounding workflow is completed.
Convert a meeting audio file into a transcript bundle, then use meeting-grounding to produce structured meeting grounding outputs.
| name | archive-grounding |
| description | Unpack a ZIP archive, inventory its files, run the corresponding child grounding skill for each supported child file, and then write a real archive-level grounded.md. |
| allowed-tools | Bash, Read, Write, Edit, Grep, Glob |
This skill handles ZIP archives as container inputs in the any-input -> grounding -> downstream research / summary / report pipeline.
Its job is not to directly pretend that the whole archive has already been deeply grounded after unpacking. Its job is to:
child_outputs/ directory,grounded.md.This skill is for archives that package together multiple materials, such as:
This skill is not a replacement for the child grounding skills themselves.
document-grounding still handles document contenttable-grounding still handles spreadsheets / CSV tablespptx-grounding still handles PowerPoint decksmeeting-audio-grounding handles meeting audio inputsmeeting-video-grounding handles meeting videosaudio_structuring remains the atomic audio transcription backend reused by the higher-level audio/video meeting entry skillsThis skill acts as an orchestrator over packaged files.
The first version should route supported child files using a simple extension-based mapping:
.pdf, .docx, .md, .txt -> document-grounding.xlsx, .csv -> table-grounding.pptx -> pptx-grounding.mp3, .wav, .m4a -> meeting-audio-grounding.mp4, .mov, .mkv -> meeting-video-groundingUnsupported files may be recorded and skipped, but they must not be silently treated as grounded.
When using this skill, you must follow this workflow:
First run the existing script entrypoint:
bash .cursor/skills/archive-grounding/scripts/run.sh <input_zip> <output_root>
The script generates an archive bundle skeleton and inventory files such as:
extracted.mdextracted_meta.jsonmanifest.jsonrouted_items.jsonunpacked/child_outputs/After the archive bundle is generated, read:
extracted.mdextracted_meta.jsonmanifest.jsonrouted_items.jsonThen enumerate the supported child files in the archive.
For each supported child file, you must actually run the corresponding child skill. Do not stop after inventory / route planning.
For child files inside the archive, all downstream child grounding outputs must be written inside the current archive bundle's child_outputs/ directory, not into the global grounding root.
Use the recommended child output path recorded in routed_items.json when present.
A child file is only considered completed if one of the following is true:
child_outputs/, orDo not pretend that a child file was fully grounded merely because it was detected, inventoried, or assigned a route.
Only after the supported child files have been processed (or explicit failures have been recorded) may you write the archive-level grounded.md.
Do not create a placeholder grounded.md.
The task is not complete if:
manifest.json / routed_items.json exist but no child skills were actually run,child_outputs/, orgrounded.md was written before the child processing step was completed.bash .cursor/skills/archive-grounding/scripts/run.sh <input_zip> <output_root>
input_zip: path to a .zip archiveoutput_root: parent directory under which the archive bundle should be createdExample:
bash .cursor/skills/archive-grounding/scripts/run.sh \
/path/to/materials.zip \
data/grounded_notes
This will create something like:
/data/grounded_notes/archive-materials/
The script should create:
<archive_bundle>/
├─ extracted.md
├─ extracted_meta.json
├─ manifest.json
├─ routed_items.json
├─ unpacked/
├─ child_outputs/
└─ grounded.md # written later by the agent, not by the script
extracted.mdHuman-readable archive overview for the agent.
It should include:
child_outputs/extracted_meta.jsonGlobal archive metadata, such as:
manifest.jsonMachine-readable file inventory for unpacked contents.
Each item should include:
routed_items.jsonMachine-readable routing plan for supported child files.
Each routed item should include:
child_outputs/Container directory for all downstream child grounding bundles generated from supported child files in this archive.
Child grounding outputs must be stored here rather than in the global grounding root.
After the child files have been processed and the child bundles exist, the agent should write a stable archive grounding note such as:
# Archive Grounding
## 1. Archive Overview
## 2. Included Materials
## 3. Successfully Processed Child Items
## 4. Key Signals Across Materials
## 5. Skipped / Unsupported / Failed Items
## 6. Suggested Next Steps
## 7. Search Keywords
This is an archive-level grounding note, not a polished final report.
Do not treat the absence of a file type as a risk by default.
Examples:
Only report:
A good result means:
child_outputs/grounded.md is written only after child processing is completeA bad result means:
grounded.md without running the child skillsgrounded.md was createdIf a supported child file cannot be processed successfully, record that explicitly.
Examples:
Do not hide child-processing failures behind a fake “archive success”.