⏳ This skill is pending AI review.

Scores will appear once the review pipeline completes.

v1.0.0

trust-portal

maxgood-ai@maxgood-ai⭐ 0 stars· last commit 6mo ago· 0 open issues

Let AI agents drive SOC 2 Type 2 compliance end-to-end. Use when the user wants to get SOC 2 compliant, start a compliance program, manage controls, view or submit evidence, check compliance scores, generate policies, review systems/vendors/risks, query the audit log, upload decision logs, manage portal settings, or configure and run automated evidence collectors (AWS, Git/CodeCommit, Platform, Policy, Vendor). TRIGGER on any of these phrases — "get me SOC 2 compliant", "set up compliance", "start our compliance program", "check compliance", "what evidence is missing", "show the audit log", "set up evidence collection", "configure collectors", "run the collectors", or any mention of SOC 2 controls, tests, evidence, policies, or evidence collectors. For end-to-end SOC 2 journey guidance, read references/soc2-playbook.md which contains the complete 8-phase workflow with exact API calls, decision trees, and conversational scripts.

Choose how to use this skill

You do not need every option. Choose the path your AI client supports. The stable page stays the same; versioned files are immutable.

1. Native installer

This listing has no registered native installer command. Use the complete package or source fallback below, depending on what your client supports.

Do not guess an installer command or replace an existing version without reviewing the diff.

2. Complete package recommended

Download the ZIP when available. It includes SKILL.md plus the references, security notes and version metadata.

No complete ProSkills package is published for this listing yet.

3. Prompt-only

Copy the prompt above when the agent can read the stable page or when you want to adopt the workflow without installing a skill.

Need only the instruction file?

Download SKILL.md only if your client requires a single file. The complete ZIP is safer for a full installation because it preserves the references and release context.

No path installs or executes anything by itself. Your agent still needs access to the project files. Before updating, compare the installed version and review the diff.

—/10

// RATINGS

⭐GitHub Stars
⭐ 0 on GitHubGitHub ↗

New / niche

🟢ProSkills Score
—
📍

Not yet listed on ClawHub or SkillsMP

// README

AI Agent-First Trust Portal — Claude Code Skill

A Claude Code agent skill for managing SOC 2 compliance data through the AI Agent-First Trust Portal API.

Features

  • Compliance status: Overall and per-category scores, evidence gaps
  • Full CRUD: Controls, tests, policies, systems, vendors, evidence, risks
  • Audit log: Query change history with filters
  • Decision logs: Upload session transcripts for compliance audit trail
  • Portal settings: View and update portal configuration
  • Pentest findings: View security assessment results
  • Evidence collectors: Configure, test, run, and inspect the portal's five evidence collectors (AWS, Git/CodeCommit, Platform, Policy, Vendor)

Installation

  1. Clone this repo into your development directory:

    git clone https://github.com/MaxGood-AI/ai-agent-first-trust-portal-skill.git
    
  2. Create a symlink in your Claude Code skills directory:

    ln -sf /path/to/ai-agent-first-trust-portal-skill ~/.claude/skills/trust-portal
    
  3. Add your trust portal credentials to .env in your workspace:

    TRUST_PORTAL_API_URL=http://localhost:5100
    TRUST_PORTAL_API_KEY=your-api-key-here
    

Usage

# Check compliance
python3 scripts/trust_portal_api.py compliance-score

# Find evidence gaps
python3 scripts/trust_portal_api.py evidence-gaps

# List controls
python3 scripts/trust_portal_api.py controls

# Query audit log
python3 scripts/trust_portal_api.py audit-log --table controls --limit 5

# Configure and run an evidence collector (credentials via data file only)
cat > /tmp/policy-config.json <<'EOF'
{"credential_mode": "none", "schedule_cron": "0 6 * * 1", "enabled": true}
EOF
python3 scripts/trust_portal_api.py configure-collector --name policy --data-file /tmp/policy-config.json
python3 scripts/trust_portal_api.py run-collector --name policy

See SKILL.md for the complete command reference.

Requirements

  • Python 3.8+ (stdlib only — no pip install needed)
  • A running AI Agent-First Trust Portal instance
  • API key from the trust portal admin (/admin/team)

License

MIT — see LICENSE.txt

// HOW IT'S BUILT

// FILES

2 files shipped with this skill. Inline text is the catalog copy. The upstream link is the source repository.

  • SKILL.md7.9 KBupstream ↗
    ---
    name: trust-portal
    description: Let AI agents drive SOC 2 Type 2 compliance end-to-end. Use when the user wants to get SOC 2 compliant, start a compliance program, manage controls, view or submit evidence, check compliance scores, generate policies, review systems/vendors/risks, query the audit log, upload decision logs, manage portal settings, or configure and run automated evidence collectors (AWS, Git/CodeCommit, Platform, Policy, Vendor). TRIGGER on any of these phrases — "get me SOC 2 compliant", "set up compliance", "start our compliance program", "check compliance", "what evidence is missing", "show the audit log", "set up evidence collection", "configure collectors", "run the collectors", or any mention of SOC 2 controls, tests, evidence, policies, or evidence collectors. For end-to-end SOC 2 journey guidance, read references/soc2-playbook.md which contains the complete 8-phase workflow with exact API calls, decision trees, and conversational scripts.
    license: MIT
    compatibility: Requires python3 and environment variables TRUST_PORTAL_API_URL and TRUST_PORTAL_API_KEY
    metadata:
      version: "1.0.0"
      openclaw:
        requires:
          env:
            - TRUST_PORTAL_API_URL
            - TRUST_PORTAL_API_KEY
          bins:
            - python3
        primaryEnv: TRUST_PORTAL_API_KEY
        homepage: https://github.com/MaxGood-AI/ai-agent-first-trust-portal
    ---
    
    # Trust Portal
    
    Manage SOC 2 compliance data through the AI Agent-First Trust Portal API.
    
    ## Environment Setup
    
    The CLI script reads two environment variables: `TRUST_PORTAL_API_URL` and `TRUST_PORTAL_API_KEY`.
    
    The script **automatically loads** these from a `.env` file if they are not already in the environment. It searches for `.env` in the current working directory and in the skill's parent directory. No shell export or command substitution is needed — just run the Python command directly.
    
    **Get your API key:** Log in to the trust portal admin at `/admin/team` and create a team member (human or agent role). The API key is shown on creation.
    
    ## Quick Start
    
    ```bash
    # Check connectivity
    python3 scripts/trust_portal_api.py health
    
    # View compliance status
    python3 scripts/trust_portal_api.py compliance-score
    
    # List controls
    python3 scripts/trust_portal_api.py controls
    
    # Find tests needing evidence
    python3 scripts/trust_portal_api.py evidence-gaps
    
    # Submit evidence
    python3 scripts/trust_portal_api.py submit-evidence \
      --test-record-id abc123 --evidence-type link \
      --url "https://..." --description "Monthly scan results"
    
    # Query audit log
    python3 scripts/trust_portal_api.py audit-log --table controls --limit 10
    
    # Upload decision log transcript
    python3 scripts/trust_portal_api.py upload-decision-log --file /path/to/transcript.jsonl
    ```
    
    ## SOC 2 Journey — End-to-End Orchestration
    
    For the complete 8-phase SOC 2 journey (bootstrap → discovery → policies → controls → evidence → gap analysis → audit prep → ongoing), read the **[SOC 2 Agent Playbook](./references/soc2-playbook.md)**. It contains exact API call sequences, decision trees, and conversational scripts for driving the entire compliance program.
    
    **Quick start:**
    1. `compliance-journey` — Check current phase and what to do next
    2. Read `references/soc2-playbook.md` for the complete workflow
    3. Use `references/discovery-questions.json` for structured discovery interviews
    
    **Resuming after interruption:** Always call `compliance-journey` first. It returns the current phase, what's completed, and the next 1-3 actions to take.
    
    ## Core Workflows
    
    ### View Compliance Status
    1. Run `compliance-score` to see overall and per-category scores.
    2. Run `evidence-gaps` to find tests needing evidence.
    
    ### Manage Controls
    1. `controls` — list all controls
    2. `control --id X` — get details for a specific control
    3. `create-control --name "..." --category security` — create a new control
    4. `update-control --id X --data-file /tmp/update.json` — update fields
    5. `delete-control --id X` — remove a control
    
    ### Record Test Execution
    1. `record-execution --test-id X --outcome success --finding "All checks passed"`
    2. `record-execution --test-id X --outcome failure --finding "MFA not enforced" --comment "2 users affected"`
    3. `record-execution --test-id X --outcome success --finding "Verified" --evidence-file /tmp/evidence.json` — include evidence
    4. `execution-history --test-id X` — view past executions for a test
    5. `execution-history --test-id X --limit 5` — limit results
    
    The `--evidence-file` flag accepts a JSON file containing an array of evidence items:
    ```json
    [
      {"evidence_type": "screenshot", "description": "Password manager showing AWS creds", "url": "https://..."},
      {"evidence_type": "link", "description": "Audit log export", "url": "https://..."},
      {"evidence_type": "file", "description": "Exported audit CSV", "file": "/path/to/audit.csv"}
    ]
    ```
    Items with a `file` field are read from disk and uploaded directly to the database. Items with a `url` field are stored as link references. Evidence items are linked to the test record and its evidence status is set to "submitted".
    
    ### Batch Operations
    1. `batch-record-execution --data-file /tmp/executions.json` — record results for multiple tests
    2. `batch-submit-evidence --data-file /tmp/evidence.json` — submit evidence to multiple tests
    
    Batch execution file format:
    ```json
    {"executions": [
      {"test_id": "abc", "outcome": "success", "finding": "All good"},
      {"test_id": "def", "outcome": "failure", "finding": "MFA not enforced", "evidence": [
        {"evidence_type": "screenshot", "description": "MFA settings page", "file": "/path/to/screenshot.png"}
      ]}
    ]}
    ```
    
    Batch evidence file format:
    ```json
    {"evidence": [
      {"test_record_id": "abc", "evidence_type": "link", "description": "Scan report", "url": "https://..."},
      {"test_record_id": "def", "evidence_type": "file", "description": "Export", "file": "/path/to/export.csv"}
    ]}
    ```
    
    Both return per-item results with succeeded/failed counts. Items with `file` paths are auto-encoded.
    
    ### Submit Evidence
    1. `evidence-gaps` — find what needs evidence
    2. `submit-evidence --test-record-id X --evidence-type link --url "..." --description "..."` — link evidence
    3. `submit-evidence --test-record-id X --evidence-type file --file /path/to/export.pdf --description "..."` — upload a file directly
    
    ### Decision Logs
    1. `upload-decision-log --file /path/to/transcript.jsonl` — upload a session transcript
    2. `decision-log-sessions` — list all uploaded sessions
    3. `decision-log-session --id X` — view session entries
    
    ### Audit Log
    1. `verify-audit-log` — verify the hash chain integrity (tamper detection)
    2. `audit-log` — recent changes across all tables
    2. `audit-log --table controls` — changes to controls only
    3. `audit-log --action DELETE --limit 5` — recent deletions
    
    ### Portal Settings
    1. `settings` — view current portal settings
    2. `update-settings --data-file /tmp/settings.json` — update settings (admin only)
    
    ### Evidence Collectors
    
    Evidence collectors are the core of SOC 2 Phase 5 (Evidence Collection). They
    gather compliance evidence from infrastructure on a schedule. The portal
    ships five: `aws`, `git`, `platform`, `policy`, `vendor`.
    
    ```bash
    # Discover which AWS account/region the portal is running in
    python3 scripts/trust_portal_api.py collector-environment
    
    # Inventory / status
    python3 scripts/trust_portal_api.py collectors
    python3 scripts/trust_portal_api.py collector --name aws
    
    # Configure a collector — credentials MUST come from a JSON file, never CLI args
    cat > /tmp/aws-config.json <<'EOF'
    {
      "credential_mode": "task_role_assume",
      "credentials": {
        "role_arn": "arn:aws:iam::123456789012:role/trust-portal-collector-role"
      },
      "config": {"region": "ca-central-1"},
      "schedule_cron": "0 6 * * 1",
      "enabled": true
    }
    EOF
    python3 scripts/trust_portal_api.py configure-collector --name aws --data-file /tmp/aws-config.json
    
    # Verify the credentials work (STS identity + optional permission probe)
    python3 scripts/trust_portal_api.py test-collector-connect
  • README.mdupstream ↗

    Contents are not inlined in the catalog. Open the upstream file.

Plain-text copy, no login: /skill-bundles/maxgood-ai-ai-agent-first-trust-portal-skill.md

// REPO STATS

0 stars
0 open issues
Last commit: 6mo ago