Social intake orchestrator
Skill Sheshiyer/skill-clusters/skills/social-intake-orchestrator
Hub-and-spoke agent-skill clusters, one per stack (Astro·GSAP·Remotion, Tauri, …). Installable via skills.sh.
npx -y skills add Sheshiyer/skill-clusters --skill social-intake-orchestratorAssembled from the repository path, not quoted from the project. Check it against their README if it does not work.
One thing to look at
- 0 stars0 stars. Stars are a popularity signal and not a quality one, but at this level it is likely that nobody has read this closely except its author, and you would be relying on your own review.
What its author says it does
Copied from the file, not written here
Automated social bookmark ingestion pipeline. Fetches Twitter/X bookmarks (bird-cli) and Instagram saved posts (gram-cli), deduplicates, enriches with AI analysis, and routes into PARA vault as Obsidian markdown. Use when asked to ingest bookmarks, sync social saves, or run the social intake pipeline.
SKILL.md
2.4 KB, 508 tokens by cl100k_base, as published. Nobody here has run it
Social Intake Orchestrator
Overview
Pulls saved/bookmarked content from Twitter/X and Instagram, enriches it with AI-generated summaries, topic tags, and Enneagram resonance analysis, then routes each item into the correct PARA vault folder as a fully-formed Obsidian markdown note.
Pipeline Stages
- FETCH — Pull new bookmarks from bird-cli and gram-cli
- NORMALIZE — Convert platform-specific JSON to unified SocialItem schema
- DEDUPE — SHA-256 hash check against registry to skip already-ingested items
- ENRICH — AI analysis: summary, topic tags, Enneagram type, PARA folder recommendation
- ROUTE — Map to correct PARA vault folder with fallback to Social-Inbox
- WRITE — Create Obsidian .md file with frontmatter, insert MOC backlinks
- STATUS — Write pipeline.json for status/monitoring
Trigger
Invoke this skill when:
- User asks to "ingest bookmarks", "sync social saves", "pull bookmarks"
- Scheduled task fires on your chosen daily cadence
- User clicks "Hunt" in the monitoring applet
Execution
Run the orchestrator:
bun <skills-dir>/social-intake-orchestrator/lib/orchestrator.ts
The orchestrator chains all pipeline stages sequentially, writes status updates at each stage, and reports results.
Configuration
- Vault path:
<VAULT>/(set to your vault root) - Hash registry:
{vault}/.social-intake-hashes.json(gitignored) - Cursor store:
<STATE_DIR>/cursors.json - Pipeline status:
<STATE_DIR>/pipeline.json - Auth: bird uses Firefox profile extraction, gram uses
~/.config/gram/config.json5
Auth Prerequisites
Before running, verify both CLIs are authenticated:
bird check # Should show ✅ auth_token and ✅ ct0
gram check # Should show "OK Credentials valid"
Manual Invocation
To run a single ingestion cycle manually:
# Full pipeline
bun <skills-dir>/social-intake-orchestrator/lib/orchestrator.ts
# With fetch limit
FETCH_LIMIT=10 bun <skills-dir>/social-intake-orchestrator/lib/orchestrator.ts
What ships with it: 14 files
49.9 KB alongside SKILL.md, 14 of them executable
lib/
- cluster.tsruns4.3 KB
- decay.tsruns4.2 KB
- dedupe.tsruns1.3 KB
- enrich.tsruns3.7 KB
- fetch.tsruns4.1 KB
- harpoon-actions.tsruns5.6 KB
- normalize.tsruns3.0 KB
- opportunity.tsruns4.3 KB
- orchestrator.tsruns4.5 KB
- status.tsruns2.5 KB
- thread-expand.tsruns3.2 KB
- transcript.tsruns3.5 KB
- types.tsruns1.1 KB
- write.tsruns4.7 KB