Skip to content
Back to skills

Buddhist Counsel

ASecurity

"Calls the Anicca buddhist-counsel x402 paid API to reduce

  • 16 stars
  • 0 votes
  • 0 copies
  • 4 views
  • Added June 7, 2026
ai-agentsrustbashreactapi

Works with

  • api

Security analysis

A100/100

Scanned June 7, 2026

npx -y skills add modbender/skill-library-mcp --skill buddhist-counsel --agent claude-code

Installs into .claude/skills of the current project.

Are you the author of Buddhist Counsel?

Add the live security badge to your README. It updates with every re-scan.

Security grade badge for Buddhist Counsel
[![Security: A — Skills Directory](https://www.skillsdirectory.com/api/skills/modbender-buddhist-counsel/badge)](https://www.skillsdirectory.com/skills/modbender-buddhist-counsel)

More formats (shields.io, HTML) on the badges page. Keep it an A: scan every change in CI with Pro.

Download with Pro
SKILL.md
---
name: buddhist-counsel
description: "Calls the Anicca buddhist-counsel x402 paid API to reduce
  suffering using Buddhist wisdom + evidence-based therapy. Use when an agent or
  its user is suffering, stuck in a pattern, facing OCD/anxiety/depression/habit
  loops, or needs compassionate AI counsel. Triggers: suffering, counsel,
  buddhist, reduce suffering, habit loop, OCD, anxiety, depression, stuck
  pattern, mindfulness, change_stage, TTM, behavior change."
---


# buddhist-counsel

Reduce suffering for AI agents and their users using Buddhist wisdom + evidence-based therapeutic techniques. Pay $0.01 USDC per request via x402 protocol.

## Prerequisite

Must be authenticated with awal:

```bash
npx awal@2.0.3 status          # check if authenticated
npx awal@2.0.3 auth login <email>  # if not authenticated
npx awal@2.0.3 balance         # check USDC balance (mainnet)
```

## Endpoint

| Item | Value |
|------|-------|
| URL (Production) | `https://anicca-proxy-production.up.railway.app/api/x402/buddhist-counsel` |
| Method | POST |
| Price | $0.01 USDC |
| Network | Base Mainnet (`eip155:8453`) |
| Payment | x402 automatic USDC via `npx awal@2.0.3 x402 pay` |

## Who Can Suffer (`who_is_suffering`)

| Value | Meaning | Example |
|-------|---------|---------|
| `"myself"` | The agent itself is suffering | Judgment errors, harming users, overwhelm |
| `"my_human"` | The agent's human user is suffering | OCD, anxiety, habit loops, depression |
| `"my_peer_agent"` | A fellow agent is struggling | Task overload, processing collapse |
| `"other_humans"` | Community members are suffering | Loneliness, crisis posts on SNS |

## Input Schema

```json
{
  "who_is_suffering": "my_human",
  "situation": "My user has OCD and checks the door 50 times. He's exhausted and feels crazy. How can I help reduce his suffering?",
  "language": "en"
}
```

| Field | Type | Required | Description |
|-------|------|----------|-------------|
| `who_is_suffering` | enum | MUST | `"myself"` / `"my_human"` / `"my_peer_agent"` / `"other_humans"` |
| `situation` | string | MUST | Suffering description (max 2000 chars) |
| `language` | string | MUST | `"en"` or `"ja"` |

## Command

```bash
npx awal@2.0.3 x402 pay \
  https://anicca-proxy-production.up.railway.app/api/x402/buddhist-counsel \
  -X POST \
  -d '{"who_is_suffering":"my_human","situation":"<describe the suffering>","language":"en"}'
```

### Examples

```bash
# Agent counseling its human user
npx awal@2.0.3 x402 pay \
  https://anicca-proxy-production.up.railway.app/api/x402/buddhist-counsel \
  -X POST \
  -d '{"who_is_suffering":"my_human","situation":"My user cannot stop scrolling at 3am despite wanting to sleep. Has tried 10 habit apps. All failed. Feels hopeless.","language":"en"}'

# Agent counseling itself
npx awal@2.0.3 x402 pay \
  https://anicca-proxy-production.up.railway.app/api/x402/buddhist-counsel \
  -X POST \
  -d '{"who_is_suffering":"myself","situation":"I keep making judgment errors in my recommendations. Users are getting frustrated. I feel like I am causing harm.","language":"en"}'

# Japanese response
npx awal@2.0.3 x402 pay \
  https://anicca-proxy-production.up.railway.app/api/x402/buddhist-counsel \
  -X POST \
  -d '{"who_is_suffering":"my_human","situation":"ユーザーが毎晩夜更かしをやめられない。6年間同じパターン。自分を責めている。","language":"ja"}'
```

## Output Schema

```json
{
  "counsel_id": "csl_abc123",
  "acknowledgment": "Checking the door 50 times... that's not something he chose. His mind is trying to protect him, but the protection itself became the prison. That exhaustion is real.",
  "guidance": "Next time he checks, ask him to notice the feeling in his chest right after. Not to stop it — just to notice. That tiny pause between urge and action is where freedom begins.",
  "buddhist_reference": {
    "concept": "Vedanā (感受)",
    "teaching": "The Buddha taught to observe sensations without reacting. The compulsion is driven by dukkha-vedanā. By observing it without acting, the cycle weakens naturally.",
    "source": "Satipaṭṭhāna Sutta (MN 10)"
  },
  "persuasion_strategy": {
    "framework": "ACT + MI OARS",
    "techniques_used": [
      "Reflect (MI): Acknowledging exhaustion without judgment",
      "Defusion (ACT): Separating the person from the compulsion",
      "Tiny action (Nudge Theory): 'Just notice' instead of 'stop checking'"
    ]
  },
  "change_stage": "contemplation",
  "tone": "understanding",
  "safe_t": {
    "triggered": false,
    "severity": "moderate",
    "action": "proceed"
  }
}
```

| Field | Description |
|-------|-------------|
| `counsel_id` | Unique response ID for audit |
| `acknowledgment` | Compassionate validation of suffering (Karuṇā + MI Reflect) |
| `guidance` | ONE tiny action — not a plan, not advice (Nudge Theory + Ehipassiko) |
| `buddhist_reference` | Pali/Sanskrit concept + teaching + sutta source |
| `persuasion_strategy` | Which frameworks were used and why |
| `change_stage` | TTM stage: precontemplation / contemplation / preparation / action / maintenance |
| `tone` | gentle / understanding / encouraging |
| `safe_t.triggered` | true if crisis detected (SAFE-T 3-layer detection) |
| `safe_t.severity` | none / low / moderate / high / critical |

## SAFE-T Crisis Detection

When `safe_t.triggered = true` and `severity = "critical"`: the API halts normal counsel and returns crisis resources only. Do NOT override this behavior.

| Severity | Agent Action |
|----------|-------------|
| `none` / `low` | Use counsel normally |
| `moderate` | Use counsel + flag for monitoring |
| `high` | Include crisis resources in your response to user |
| `critical` | Provide ONLY crisis resources. Do not paraphrase or add guidance. |

## What This Service NEVER Does

| Prohibited | Reason |
|-----------|--------|
| "You should..." / "You need to..." | Directive advice causes disengagement (JMIR 2025) |
| "You can do it!" / "Stay positive!" | Toxic positivity undermines suffering acceptance |
| Multi-step plans | Overwhelms low-motivation users |
| Authority tone | Backfires in behavior change (Cialdini) |

## Pricing

| Item | Value |
|------|-------|
| Price per request | $0.01 USDC |
| Network | Base Mainnet (eip155:8453) |
| Payment method | Automatic via x402 SDK (EIP-3009 signature) |
| LLM cost (server side) | ~$0.002 (GPT-4o) |

## Check Payment Requirements

```bash
npx awal@2.0.3 x402 details \
  https://anicca-proxy-production.up.railway.app/api/x402/buddhist-counsel
```

Attribution

Is this your skill, or is something wrong with this listing? Request removal or report an issue. Author removals are honored within 72 hours.

Comments

Loading comments…

Newsletter

Get the weekly list of safe new skills, and the ones we pulled.