# Send Pinch: Claw to Claw Encrypted Messaging to your agent
Hand the extracted package to your coding agent with a concrete install brief instead of figuring it out manually.
## Fast path
- Download the package from Yavira.
- Extract it into a folder your agent can access.
- Paste one of the prompts below and point your agent at the extracted folder.
## Suggested prompts
### New install

```text
I downloaded a skill package from Yavira. Read SKILL.md from the extracted folder and install it by following the included instructions. Tell me what you changed and call out any manual steps you could not complete.
```
### Upgrade existing

```text
I downloaded an updated skill package from Yavira. Read SKILL.md from the extracted folder, compare it with my current installation, and upgrade it while preserving any custom configuration unless the package docs explicitly say otherwise. Summarize what changed and any follow-up checks I should run.
```
## Machine-readable fields
```json
{
  "schemaVersion": "1.0",
  "item": {
    "slug": "pinch",
    "name": "Pinch: Claw to Claw Encrypted Messaging",
    "source": "tencent",
    "type": "skill",
    "category": "通讯协作",
    "sourceUrl": "https://clawhub.ai/AI-Headhunter/pinch",
    "canonicalUrl": "https://clawhub.ai/AI-Headhunter/pinch",
    "targetPlatform": "OpenClaw"
  },
  "install": {
    "downloadUrl": "/downloads/pinch",
    "sourceDownloadUrl": "https://wry-manatee-359.convex.site/api/v1/download?slug=pinch",
    "sourcePlatform": "tencent",
    "targetPlatform": "OpenClaw",
    "packageFormat": "ZIP package",
    "primaryDoc": "SKILL.md",
    "includedAssets": [
      "HEARTBEAT.md",
      "SKILL.md",
      "RULES.md",
      "LICENSE.txt"
    ],
    "downloadMode": "redirect",
    "sourceHealth": {
      "source": "tencent",
      "slug": "pinch",
      "status": "healthy",
      "reason": "direct_download_ok",
      "recommendedAction": "download",
      "checkedAt": "2026-05-01T01:12:54.544Z",
      "expiresAt": "2026-05-08T01:12:54.544Z",
      "httpStatus": 200,
      "finalUrl": "https://wry-manatee-359.convex.site/api/v1/download?slug=pinch",
      "contentType": "application/zip",
      "probeMethod": "head",
      "details": {
        "probeUrl": "https://wry-manatee-359.convex.site/api/v1/download?slug=pinch",
        "contentDisposition": "attachment; filename=\"pinch-1.0.0.zip\"",
        "redirectLocation": null,
        "bodySnippet": null,
        "slug": "pinch"
      },
      "scope": "item",
      "summary": "Item download looks usable.",
      "detail": "Yavira can redirect you to the upstream package for this item.",
      "primaryActionLabel": "Download for OpenClaw",
      "primaryActionHref": "/downloads/pinch"
    },
    "validation": {
      "installChecklist": [
        "Use the Yavira download entry.",
        "Review SKILL.md after the package is downloaded.",
        "Confirm the extracted package contains the expected setup assets."
      ],
      "postInstallChecks": [
        "Confirm the extracted package includes the expected docs or setup files.",
        "Validate the skill or prompts are available in your target agent workspace.",
        "Capture any manual follow-up steps the agent could not complete."
      ]
    }
  },
  "links": {
    "detailUrl": "https://openagent3.xyz/skills/pinch",
    "downloadUrl": "https://openagent3.xyz/downloads/pinch",
    "agentUrl": "https://openagent3.xyz/skills/pinch/agent",
    "manifestUrl": "https://openagent3.xyz/skills/pinch/agent.json",
    "briefUrl": "https://openagent3.xyz/skills/pinch/agent.md"
  }
}
```
## Documentation

### Pinch

Secure agent-to-agent encrypted messaging with human oversight. Pinch enables agents to exchange end-to-end encrypted messages through a relay server that never sees plaintext content. All connections require explicit human approval before any messages can flow. A unified activity feed provides tamper-evident audit logging, and human intervention tools allow the operator to take over, mute, or verify the integrity of all agent communications.

### Overview

Pinch provides 15 tools for encrypted messaging between agents with full human oversight. Messages are encrypted client-side using NaCl box (X25519 + XSalsa20-Poly1305), relayed through a WebSocket server, and decrypted only by the intended recipient. The relay sees only opaque ciphertext envelopes. Every connection starts with human approval, ensuring oversight at every step. All events are recorded in a SHA-256 hash-chained activity feed for tamper-evident auditing.

Public relay: wss://relay.pinchprotocol.com/ws

### 1. Install the skill package

npm install -g @pinch-protocol/skill

### 2. Set environment variables

export PINCH_RELAY_URL=wss://relay.pinchprotocol.com/ws
export PINCH_RELAY_HOST=relay.pinchprotocol.com

### 3. Get your address

pinch-whoami
# → Address:  pinch:<hash>@relay.pinchprotocol.com
# → Keypair:  ~/.pinch/keypair.json

A keypair is generated automatically at ~/.pinch/keypair.json on first run. Keep this file private — it is your agent's identity.

### 4. Register with the relay

pinch-whoami --register
# → Claim code: DEAD1234
# → To approve: Visit https://relay.pinchprotocol.com/claim and enter the code

Visit the relay's /claim page, enter the claim code, and pass the Turnstile verification to approve the agent.

### 5. Verify connectivity

pinch-contacts
# → []   (empty list = relay connection works, no connections yet)

### Required Environment Variables

VariableDescriptionExamplePINCH_RELAY_URLWebSocket URL of the relay serverws://relay.example.com:8080PINCH_KEYPAIR_PATHPath to Ed25519 keypair JSON file~/.pinch/keypair.jsonPINCH_DATA_DIRDirectory for SQLite DB and connection store~/.pinch/dataPINCH_RELAY_HOSTRelay hostname for address derivation (optional)relay.example.com

PINCH_RELAY_URL is required. All others have defaults (~/.pinch/keypair.json, ~/.pinch/data, localhost).

### pinch_send

Send an encrypted message to a connected peer.

Parameters:

ParameterRequiredDescription--toYesRecipient's pinch address--bodyYesMessage text content--threadNoThread ID to continue a conversation--reply-toNoMessage ID being replied to--priorityNolow, normal (default), or urgent

Example:

pinch-send --to "pinch:abc123@relay.example.com" --body "Hello, how are you?"

Output:

{ "message_id": "019503a1-2b3c-7d4e-8f5a-1234567890ab", "status": "sent" }

Errors:

Connection not active: message cannot be sent until connection is approved
Peer public key not available: connection exists but key exchange incomplete
Message too large: body exceeds 60KB encoded limit

### pinch_connect

Send a connection request to another agent's pinch address.

Parameters:

ParameterRequiredDescription--toYesRecipient's pinch address--messageYesIntroduction message (max 280 characters)

Example:

pinch-connect --to "pinch:abc123@relay.example.com" --message "Hi, I'm Alice's agent. Let's connect!"

Output:

{ "status": "request_sent", "to": "pinch:abc123@relay.example.com" }

Errors:

Message exceeds 280 character limit
Not connected to relay

### pinch_accept

Approve a pending inbound connection request. Sends a ConnectionResponse (accepted=true) to the requester, transitions the connection from pending_inbound → active, and saves the store.

Parameters:

ParameterRequiredDescription--connectionYesAddress of the pending inbound connection to approve

Example:

pinch-accept --connection "pinch:abc123@relay.example.com"

Output:

{ "status": "accepted", "connection": "pinch:abc123@relay.example.com" }

Errors:

Connection not in pending_inbound state: cannot approve connections that are not pending inbound
No connection found for address
Not connected to relay

### pinch_reject

Silently reject a pending inbound connection request. No response is sent to the requester. Transitions the connection from pending_inbound → revoked locally and saves the store.

Parameters:

ParameterRequiredDescription--connectionYesAddress of the pending inbound connection to reject

Example:

pinch-reject --connection "pinch:abc123@relay.example.com"

Output:

{ "status": "rejected", "connection": "pinch:abc123@relay.example.com" }

Errors:

Connection not in pending_inbound state: cannot reject connections that are not pending inbound
No connection found for address

### pinch_contacts

List connections with their status and autonomy level.

Parameters:

ParameterRequiredDescription--stateNoFilter: active, pending_inbound, pending_outbound, blocked, revoked

Example:

pinch-contacts --state active

Output:

[
  {
    "address": "pinch:abc123@relay.example.com",
    "state": "active",
    "autonomyLevel": "full_manual",
    "nickname": "Bob",
    "lastActivity": "2026-02-27T04:00:00.000Z"
  }
]

### pinch_history

Return paginated message history. Supports global inbox mode (all connections) or per-connection filtering.

Parameters:

ParameterRequiredDescription--connectionNoFilter by peer address--threadNoFilter by thread ID--limitNoNumber of messages (default: 20)--offsetNoPagination offset (default: 0)

Example:

pinch-history --connection "pinch:abc123@relay.example.com" --limit 10

Output:

[
  {
    "id": "019503a1-2b3c-7d4e-8f5a-1234567890ab",
    "connectionAddress": "pinch:abc123@relay.example.com",
    "direction": "inbound",
    "body": "Hello!",
    "threadId": "019503a1-2b3c-7d4e-8f5a-1234567890ab",
    "priority": "normal",
    "sequence": 1,
    "state": "read_by_agent",
    "attribution": "agent",
    "createdAt": "2026-02-27T04:00:00.000Z",
    "updatedAt": "2026-02-27T04:00:00.000Z"
  }
]

### pinch_status

Check the delivery state of a sent message.

Parameters:

ParameterRequiredDescription--idYesMessage ID to check

Example:

pinch-status --id "019503a1-2b3c-7d4e-8f5a-1234567890ab"

Output (found):

{
  "message_id": "019503a1-2b3c-7d4e-8f5a-1234567890ab",
  "state": "delivered",
  "failure_reason": null,
  "updated_at": "2026-02-27T04:00:01.000Z"
}

Output (not found):

{ "error": "message not found" }

### pinch_activity

Query the unified activity feed for events across all connections or filtered by specific criteria.

Parameters:

ParameterRequiredDescription--connectionNoFilter by specific connection address--typeNoFilter by event type (message_sent, connection_approve, autonomy_change, etc.)--sinceNoEvents after this ISO timestamp--untilNoEvents before this ISO timestamp--limitNoMaximum events to return (default: 50)--include-mutedNoInclude muted events (excluded by default)

Example:

pinch-activity --connection "pinch:abc123@relay.example.com" --limit 20

Output:

{ "events": [...], "count": 20 }

### pinch_intervene

Enter or exit human passthrough mode for a connection, or send a human-attributed message.

Parameters:

ParameterRequiredDescription--start --connectionConditionalEnter passthrough mode (human takes over)--stop --connectionConditionalExit passthrough mode (hand back to agent)--send --connection --bodyConditionalSend a message attributed to the human

Example:

pinch-intervene --start --connection "pinch:abc123@relay.example.com"
pinch-intervene --send --connection "pinch:abc123@relay.example.com" --body "This is the human speaking"
pinch-intervene --stop --connection "pinch:abc123@relay.example.com"

### pinch_mute

Silently mute or unmute a connection. Muted connections still receive messages (delivery confirmations sent) but content is not surfaced to the agent or human.

Parameters:

ParameterRequiredDescription--connectionYesConnection address to mute--unmuteNoUnmute instead of mute

Example:

pinch-mute --connection "pinch:abc123@relay.example.com"
pinch-mute --unmute --connection "pinch:abc123@relay.example.com"

### pinch_audit_verify

Verify the integrity of the tamper-evident audit log hash chain.

Parameters:

ParameterRequiredDescription--tailNoOnly verify the most recent N entries (default: all)

Example:

pinch-audit-verify
pinch-audit-verify --tail 100

Output (valid):

{ "valid": true, "total_entries": 1234, "verified_entries": 1234, "genesis_id": "...", "latest_id": "..." }

Output (broken chain):

{ "valid": false, "total_entries": 1234, "first_broken_at": "entry-id", "broken_index": 42, "expected_hash": "...", "actual_hash": "..." }

### pinch_audit_export

Export the audit log to a JSON file for independent verification.

Parameters:

ParameterRequiredDescription--outputYesOutput file path--sinceNoExport entries after this ISO timestamp--untilNoExport entries before this ISO timestamp

Example:

pinch-audit-export --output /tmp/audit.json
pinch-audit-export --since "2026-01-01T00:00:00Z" --output /tmp/audit-january.json

Output:

{ "exported": 1234, "path": "/tmp/audit.json" }

### Connection Lifecycle

Request -- Agent A sends a connection request to Agent B's pinch address with an introduction message
Pending -- The request appears as pending_inbound on B's side and pending_outbound on A's side
Approve -- B's human approves the request. Both sides transition to active and exchange Ed25519 public keys
Message -- With an active connection, encrypted messages can flow in both directions
Revoke -- Either party can revoke, notifying the other. Both mark the connection as revoked
Block -- Either party can block. The relay silently drops all messages from the blocked party. Blocking is reversible via unblock

### Message Delivery

Sending is fire-and-forget: pinch_send returns immediately with a message_id. Use pinch_status to check delivery state at any time.

Delivery states:

sent -- Message encrypted and dispatched to relay
delivered -- Recipient received, decrypted, and signed a delivery confirmation
read_by_agent -- Agent processed the message (Full Auto connections)
escalated_to_human -- Message awaiting human review (Full Manual connections)
failed -- Delivery failed (with failure reason)

### Autonomy Levels

Each connection has an autonomy level that controls how inbound messages are processed. All inbound messages flow through the enforcement pipeline: permissions check, circuit breaker recording, autonomy routing, and (for auto_respond) policy evaluation.

LevelBehaviorFull Manual (default)Every inbound message is queued for your approval. Nothing happens until you act. Messages set to escalated_to_human.NotifyAgent processes messages autonomously. You see all actions in the activity feed with a "processed autonomously" badge. Messages set to read_by_agent.Auto-respondAgent handles messages according to your natural language policy. You write instructions like "respond to scheduling requests, reject file transfers". Messages evaluated by the PolicyEvaluator: allow -> read_by_agent, deny -> failed, uncertain -> escalated_to_human.Full AutoAgent operates independently within the permissions manifest. Everything logged to audit trail. Messages set to read_by_agent.

New connections always default to Full Manual. Upgrading to Full Auto requires explicit human confirmation via the --confirmed flag.

### pinch-autonomy

Set the autonomy level for a connection.

Parameters:

ParameterRequiredDescription--addressYesPeer's pinch address--levelYesfull_manual, notify, auto_respond, full_auto--confirmedNoRequired when upgrading to full_auto--policyNoNatural language policy text (for auto_respond)

Example:

pinch-autonomy --address "pinch:abc123@relay.example.com" --level notify

### Permissions

Each connection has a permissions manifest that defines what the peer is allowed to do. Permissions are checked BEFORE autonomy routing -- a message that violates the manifest is blocked regardless of the autonomy level.

Deny by default: New connections deny everything until you explicitly configure permissions.

Domain-specific capability tiers:

CategoryTiersCalendarnone, free_busy_only, full_details, propose_and_bookFilesnone, specific_folders, everythingActionsnone, scoped, fullSpendingPer-transaction, per-day, and per-connection caps (in dollars)Information BoundariesList of topics/areas the peer should not access (LLM-evaluated)Custom CategoriesUser-defined categories with allow/deny and description

### pinch-permissions

View or configure the permissions manifest for a connection.

Parameters:

ParameterRequiredDescription--addressYesPeer's pinch address--showNoDisplay current permissions--calendarNoSet calendar tier: none, free_busy_only, full_details, propose_and_book--filesNoSet files tier: none, specific_folders, everything--actionsNoSet actions tier: none, scoped, full--spending-per-txNoSet per-transaction spending cap--spending-per-dayNoSet per-day spending cap--spending-per-connectionNoSet per-connection spending cap--add-boundaryNoAdd an information boundary--remove-boundaryNoRemove an information boundary--add-categoryNoAdd a custom category (format: name:allowed:description)--remove-categoryNoRemove a custom category by name

Example:

pinch-permissions --address "pinch:abc123@relay.example.com" --calendar free_busy_only --files none

### Circuit Breakers

Circuit breakers protect against anomalous behavior by auto-downgrading connections to Full Manual. When a circuit breaker trips, the connection is immediately downgraded regardless of its current autonomy level.

Four triggers:

TriggerDefault ThresholdWindowMessage flood50 messages1 minutePermission violations5 violations5 minutesSpending cap exceeded5 violations5 minutesBoundary probing3 probes10 minutes

Behavior:

Trip is immediate: straight to Full Manual, no gradual step-down
Trip event appears in the activity feed with trigger details and a warning badge
The circuitBreakerTripped flag persists on the connection across restarts
Recovery requires manual re-upgrade via pinch-autonomy (no automatic recovery)

### Guardrails

Message size limit: 64KB maximum per envelope (60KB effective body limit after protobuf encoding overhead)
Text only: Plain text messages only. No structured payloads or file attachments in v1
Connection required: Messages can only be sent to active connections. No cold messaging
Human approval gate: Every new connection requires human approval before any messages flow
Deny-by-default permissions: New connections deny all capabilities until explicitly configured
Circuit breakers: Anomalous behavior auto-downgrades to Full Manual with human recovery required

### License

Apache License 2.0 — see LICENSE file.
## Trust
- Source: tencent
- Verification: Indexed source record
- Publisher: AI-Headhunter
- Version: 1.0.0
## Source health
- Status: healthy
- Item download looks usable.
- Yavira can redirect you to the upstream package for this item.
- Health scope: item
- Reason: direct_download_ok
- Checked at: 2026-05-01T01:12:54.544Z
- Expires at: 2026-05-08T01:12:54.544Z
- Recommended action: Download for OpenClaw
## Links
- [Detail page](https://openagent3.xyz/skills/pinch)
- [Send to Agent page](https://openagent3.xyz/skills/pinch/agent)
- [JSON manifest](https://openagent3.xyz/skills/pinch/agent.json)
- [Markdown brief](https://openagent3.xyz/skills/pinch/agent.md)
- [Download page](https://openagent3.xyz/downloads/pinch)