digest-bridge

name	digest-bridge
description	Use when the user wants to extract arXiv IDs or DOIs from research or RSS digests and turn them into getscipapers requests or manifests.
metadata	{"short-description":"Bridge digest outputs into paper retrieval"}

Digest Bridge

Windows Runtime Commands

On native Windows, use the managed Windows runner and the native runtime command target. Set $runtime to the installed runtime root. Multi-agent installs usually use %LOCALAPPDATA%\ai-agents-skills\runtime. Then run:

$runtime = if ($env:AAS_RUNTIME_ROOT) { $env:AAS_RUNTIME_ROOT } else { "$env:LOCALAPPDATA\ai-agents-skills\runtime" }
& "$runtime\run_skill.bat" "skills/digest-bridge/run_digest_bridge.bat" <args>

POSIX examples below use run_skill.sh and .sh command targets; use the Windows command target above on native Windows.

This uses the managed ai-agents-skills runtime copy of the digest bridge workflow.

When to use

• scan research digests for papers
• scan RSS digests for papers
• create a getscipapers manifest from digest outputs
• request papers mentioned in recent digests

Base path

• $AAS_RUNTIME_WORKSPACE/skills/digest-bridge/

This is a direct Python entry point, so run it from the managed ai-agents-skills runtime workspace with the workspace-local PYTHONPATH.

Core commands

Use functions.exec_command.

cd "$AAS_RUNTIME_WORKSPACE" && PYTHONPATH="$AAS_RUNTIME_WORKSPACE/.local:${PYTHONPATH:-}" python3 skills/digest-bridge/digest_bridge.py scan

cd "$AAS_RUNTIME_WORKSPACE" && PYTHONPATH="$AAS_RUNTIME_WORKSPACE/.local:${PYTHONPATH:-}" python3 skills/digest-bridge/digest_bridge.py scan --source research --min-score 3

cd "$AAS_RUNTIME_WORKSPACE" && PYTHONPATH="$AAS_RUNTIME_WORKSPACE/.local:${PYTHONPATH:-}" python3 skills/digest-bridge/digest_bridge.py request --source research

cd "$AAS_RUNTIME_WORKSPACE" && PYTHONPATH="$AAS_RUNTIME_WORKSPACE/.local:${PYTHONPATH:-}" python3 skills/digest-bridge/digest_bridge.py request --source rss --watch

Operational notes

• Use this after a digest run, not as a replacement for the digest itself.
• Respect --source and --min-score filters instead of broad requests when the user wants a narrower batch.
• If the user wants actual external retrieval, follow the manifest or request output into getscipapers-requester.
• scan is the dry-run discovery step; request is the transition into manifest/watch creation.