Search the documents uploaded for the CURRENT task/row (e.g. an Agentic Table row's attached files) via the `unique-cli uploaded-search "<query>"` command, with the same per-turn citation tracking as `unique-cli search` so cited facts render as `<sup>N</sup>` footnotes and clickable reference chips on the Unique platform. ALWAYS use this skill when the user refers to documents they uploaded/attached to this task, or when you need facts from the task's own attached files. These uploaded files are scoped to the chat, NOT to a knowledge-base folder, so they will NEVER appear in `unique-cli search` results no matter the folder or metadata filters. Use `unique-cli search` for the knowledge base and this command for the task's uploaded documents; the two are complementary and citation numbering is shared across them within a turn.
The English summary is being prepared. Showing the original description.
See how to installReview the instructions the agent will receive before you install.
Documents uploaded for the current task (for example an Agentic Table
row's attached files) are not part of the knowledge-base folder scope.
They are scoped to the chat, so they will never appear in
unique-cli search results — no folder or metadata filter can surface them.
Use unique-cli uploaded-search to retrieve them.
# Search the documents uploaded for this row/task
unique-cli uploaded-search "target asset classes and investment strategy"
# Limit results
unique-cli uploaded-search "fee structure" --limit 50
There is no --folder/--metadata for uploaded-search — the set of
uploaded documents is fixed by what was attached to the task.
If no documents were uploaded for this task, the command reports that and you
should fall back to unique-cli search for the knowledge base.
| You want to search… | Command |
|---|---|
| The knowledge base (folders the task scope grants) | unique-cli search "<query>" |
| Documents uploaded for this row/task | unique-cli uploaded-search "<query>" |
Each result is rendered as a <sourceN>...</sourceN> block, identical to
unique-cli search. N is 1-based and shared with unique-cli search
within the same turn — both commands append to the same per-turn citation
manifest, so numbering stays continuous across them (a search call that
emitted <source1>–<source3> is followed by an uploaded-search call whose
first result is <source4>).
Found 1 result(s):
<source1>
<|document|>investment-mandate.pdf</|document|>
<|page|>2</|page|>
<|info|>cont_abc123</|info|>
...the mandate targets EMEA equities with a 5% cap per issuer...
</source1>
Cite a fact from uploaded-search results with [sourceN], exactly as you
would for unique-cli search — the two share the same namespace and manifest.
The Unique platform converts each [sourceN] marker in your final answer into a
<sup>N</sup> footnote and a clickable reference chip.
The uploaded mandate caps single-issuer exposure at 5% [source1].
Rules (enforced by the platform's reference post-processor):
[sourceN] is for KB and uploaded-document results. Web results from
unique-cli web-search use [websourceN] — never mix the two namespaces.search /
uploaded-search output. Numbers from previous turns are stale and will be
silently dropped.source in singular form with the number in digits: [source1],
[source2] — not [Source 1] or [source one].unique-cli uploaded-search <query> [--limit <N>]
| Option | Short | Default | Description |
|---|---|---|---|
--limit | -l | 200 | Max results |
Requires these environment variables:
UNIQUE_USER_ID # User ID (required)
UNIQUE_COMPANY_ID # Company ID (required)
UNIQUE_API_KEY # API key — optional on localhost / secured cluster
UNIQUE_APP_ID # App ID — optional on localhost / secured cluster
Install: pip install unique-sdk
No reviews yet. Tell others what you thought.
No login needed. Share how it worked for you, ask a question, or note where it did not run.
No comments yet. Be the first to write one.
Implement comprehensive error handling for Python code paths to keep services resilient and user-friendly. Use when failures are currently silent or exceptions leak through.
The English summary is being prepared. Showing the original description.
Tabular and numerical data analysis with descriptive statistics and insights. Use when the user provides data, tables, CSVs, or numbers and wants analysis.
The English summary is being prepared. Showing the original description.
Financial factsheet analysis with key metrics extraction and investment rationale
The English summary is being prepared. Showing the original description.
Diagnose and fix CI failures without leaving your editor.
The English summary is being prepared. Showing the original description.
Reproduce ai-repo PR checks locally with Poe and CI scripts, including per-package typecheck and coverage behavior. Use when validating changes before push or when user asks which local commands match CI.
The English summary is being prepared. Showing the original description.
Ask clarifying questions before implementing to ensure Python requirements are understood. Use when a task lacks detail, dependencies are unclear, or multiple interpretations are possible.
The English summary is being prepared. Showing the original description.