Reconnect for Openclaw

Reconnect helps users discover relevant professional contacts, organize evidence, and manage authorised networking outreach safely across resumable sessions.

antreasantoniou
v0.1.0-rc.2
Sep 6, 2026
0
136
0

Install & Download

1. ClawHub CLI

The fastest way to install a skill directly from the registry.

npx clawhub@latest install reconnect

2. Manual Installation

Copy the skill folder to one of these locations

Global
~/.openclaw/skills/
Workspace
<project>/skills/

Priority: Workspace > Local > Bundled

3. Prompt Installation

Copy this prompt to OpenClaw to install it automatically.

Help me install reconnect using Clawhub. If Clawhub is not installed, install it first (npm i -g clawhub).

Prefer to download?

Get the raw skill files in a ZIP archive.

What is Reconnect?

Reconnect is an Openclaw Skills workflow for goal-based professional networking. It helps identify relevant peers, collaborators, mentors, alumni, colleagues, communities, and conference contacts, while preserving source evidence, identity distinctions, eligibility decisions, and outreach outcomes.

The skill supports discovery, preparation, authorised connection actions, reconciliation, and resumption. Its Python helper performs local evidence imports, destination deduplication, transactional batch reservations, observations, and JSON exports without making network calls or operating LinkedIn directly.

Reconnect Use Cases

  • Discover peers, mentors, collaborators, or communities around a professional goal, topic, role, or field.
  • Reconnect with alumni, former colleagues, existing contacts, or people associated with an event.
  • Follow up on conference, course, employer, roster, or community relationships.
  • Prepare a reviewable batch of potential LinkedIn recipients before outreach.
  • Resume interrupted networking work without losing reservations, evidence, or outcome state.
  • Reconcile uncertain browser outcomes before attempting another invitation.
  • Import permitted CRM exports, address books, event lists, or community databases through project-specific adapters.
  • Maintain privacy-conscious networking records without publishing personal data.

How Reconnect Works

  1. Determine the requested mode: Find, Prepare, Connect, or Reconcile/resume.
  2. For a new project or changed audience, read the network plan and resolve a private project directory for preferences, sources, candidates, reservations, and outcomes.
  3. In Find mode, research public sources and record candidates with source URLs, dates, supporting evidence, and conflicting evidence.
  4. In Prepare mode, refresh source state, import generic candidate JSON, preserve provenance and exclusions, deduplicate canonical destinations, and reserve a stable batch transactionally.
  5. Separate identity evidence, relevance, review priority, relationship, and group membership; surface unresolved identity or eligibility issues for review.
  6. In Connect mode, read the LinkedIn workflow, resolve each specifically authorised recipient, inspect the live profile, and perform only permitted actions using supported browser tools.
  7. Keep authenticated LinkedIn work sequential in one tab, preserve the project invitation note, tone, and sender identity, and use only truthful connection reasons.
  8. Verify the resulting state instead of treating profile views, button presses, or timeouts as successful invitations.
  9. In Reconcile/resume mode, audit uncertain outcomes, update both the canonical contact tracker and batch ledger, and stop when the platform presents warnings, limits, restrictions, or identity challenges.

Reconnect Setup

  1. Install or enable the Reconnect skill in the Openclaw Skills environment and keep the distributable skill separate from private project data.
  2. Create or select a private project directory for preferences, source records, candidate data, batch reservations, and outcome tracking.
  3. Review the referenced workflow documents before operating:
    • references/network-plan.md for project goals and audience changes.
    • references/discovery.md for research and evidence capture.
    • references/tracking.md for imports, reservations, ledgers, and helper commands.
    • references/linkedin.md for authorised LinkedIn actions.
  4. Use the bundled Python helper with the standard library only; no additional Python dependencies or network service are required.
python3 scripts/reconnect.py --help
  1. Configure any source-specific adapter in the private project. Adapters may emit the generic candidate JSON format, but must preserve provenance and existing exclusions.
  2. Use the current browser harness and supported browser tools for authenticated LinkedIn work. Do not place credentials, generated databases, batch outputs, or personal data in the installed skill, its snapshots, or public repositories.

Reconnect Data Schema & Taxonomy

Reconnect separates project state into reviewable evidence, candidate, batch, and outcome information rather than collapsing all decisions into a single contact record.

Data area Purpose
Project preferences Stores the networking goal, audience, invitation note, tone, and sender identity.
Source records Preserve source URLs, source dates, provenance, and supporting or conflicting evidence.
Candidate records Represent generic candidate JSON imported from permitted adapters or research.
Identity evidence Tracks whether a person or destination has been resolved and distinguishes archival identities.
Relevance and eligibility Keeps fit for the user's goal separate from identity confirmation and campaign eligibility.
Relationship and groups Records contact relationship and group membership independently.
Batch reservations Provides stable, transactionally reserved outreach batches for review and resumption.
Observations and outcomes Records browser observations and verified results; uncertain states remain explicitly uncertain.
Canonical contact tracker Maintains the project's authoritative contact state alongside the batch ledger.

Destinations are deduplicated by canonical destination while underlying source IDs are preserved. The helper exports reviewable JSON and makes no network calls. Generated databases and batch outputs must remain in the private project and must not be committed, pushed, or published.

Reconnect Advanced Features

  • Resumable, transactional batch reservations that survive interruptions and prevent unstable recipient selection.
  • Evidence-aware review ordering using a score that prioritizes review without representing a probability.
  • Provenance-preserving imports from project-specific CRM, address book, event, or community adapters.
  • Canonical destination deduplication while retaining multiple archival identities and source IDs.
  • Separate tracking for identity, relevance, eligibility, relationships, groups, observations, and outcomes.
  • Reconciliation workflows that distinguish successful invitations from profile opens, button presses, and timeouts.
  • Persistent user approval scope for an explicitly identified batch without repeatedly requesting the same approval.
  • Sequential authenticated LinkedIn operation in one browser tab through the active browser harness.
  • Safety controls that stop on platform warnings, rate limits, restrictions, or identity challenges.
  • Privacy-preserving project storage with local standard-library processing and reviewable JSON exports.
  • Truthful, user-specific invitation notes that preserve the selected tone and sender identity.
  • Openclaw Skills compatibility for structured professional networking workflows without claiming a universal social API.

SKILL.md


Loading

Related Openclaw Skills

METADATA

Github Stars: 0
forks: 0

Featured*