claude-autopmBLOCK
Autonomous Project Management Framework for Claude Code - Advanced AI-powered development automation
Overview
From the repository's own README, as read at the audited commit. Badges and raw HTML are left out.
[](https://www.npmjs.com/package/claude-autopm) [](https://www.npmjs.com/package/claude-autopm) [](LICENSE) [](https://github.com/rafeekpro/ClaudeAutoPM)
AI-Powered Project Management Framework for Claude Code
ClaudeAutoPM transforms your development workflow with intelligent automation, specialized AI agents via a plugin system, and complete GitHub/Azure DevOps integration. Install only what you need — pay only for the tokens you use.
Quick Install
Requires Node.js >= 22 and npm >= 10.
npm install -g claude-autopm cd your-project autopm install
Choose your scenario during installation:
Architecture
Plugin-Based Agent System
ClaudeAutoPM uses a plugin architecture for token efficiency. Only 7 core agents are loaded by default. Specialized agents are installed on demand via plugins.
autopm/.claude/agents/ ← 7 core agents (always available) packages/plugin-*/agents/ ← 50+ specialized agents (per plugin)
Token savings: Lite installs use 73% fewer agent tokens vs full.
XML Rules Engine
8 critical rules auto-loaded via @include in XML format:
48b4be9853eeOBSERVED · 2026-10-12Connect
Built from this server's own package name, version and transport as found in its source — not copied from anyone's documentation, so it cannot drift against a page we do not control. Replace the environment placeholders with a token scoped to the least it needs.
claude mcp add plugin-testing --env ANTHROPIC_API_KEY=${ANTHROPIC_API_KEY} --env AUTOPM_MASTER_PASSWORD=${AUTOPM_MASTER_PASSWORD} --env AZURE_DEVOPS_TOKEN=${AZURE_DEVOPS_TOKEN} --env CONTEXT7_API_KEY=${CONTEXT7_API_KEY} -- npx -y @claudeautopm/[email protected]{
"mcpServers": {
"plugin-testing": {
"command": "npx",
"args": [
"-y",
"@claudeautopm/[email protected]"
],
"env": {
"ANTHROPIC_API_KEY": "${ANTHROPIC_API_KEY}",
"AUTOPM_MASTER_PASSWORD": "${AUTOPM_MASTER_PASSWORD}",
"AZURE_DEVOPS_TOKEN": "${AZURE_DEVOPS_TOKEN}",
"CONTEXT7_API_KEY": "${CONTEXT7_API_KEY}"
}
}
}
}Exposed tools (27)
22 read · 5 write · 0 destructive.
| Tool | Risk | Description |
|---|---|---|
Checkout | read | Payment processing |
Dashboard | read | Main view |
F1 | read | Desc1 |
Login | read | User authentication |
autopm_checkpoint | write | Create a project checkpoint |
autopm_close_issue | read | Close an issue |
autopm_create_issue | write | Create a new local issue |
autopm_epic_template | read | Template for creating a new epic |
autopm_issue_template | read | Template for creating a new issue |
autopm_learn | write | Save a project learning |
autopm_list_epics | read | List epics with progress |
autopm_list_issues | read | List local issues with optional status filter |
autopm_list_prds | read | List PRDs |
autopm_prd_template | read | Template for creating a new PRD |
autopm_recall | read | Get project learnings |
autopm_show_epic | read | Show epic with tasks |
autopm_show_issue | read | Show issue details by ID |
autopm_show_prd | read | Show PRD details |
autopm_start_issue | write | Start working on an issue (set in_progress) |
autopm_status | read | Project overview — counts, recent activity |
aws-architect | read | AWS |
epicSync | write | Sync epic to GitHub/Azure DevOps |
issueAnalyze | read | Analyze issue for implementation approach |
new-name | read | new description |
test-agent | read | Test agent |
test-project | read | Test Project |
user-management | read | User management epic |
Trust audit
BLOCKgrade F · trust 30/100 Do not install this without reading the findings. The audit found something that could harm you or your machine.
| Layer | What it checks | Result |
|---|---|---|
| L0 | Provenance & inventory | PASS |
| L1 | Static analysis of the code | FAIL |
| L2 | Instruction surface (what it tells the agent) | FAIL |
| L3 | Class-specific surface | PASS |
| L4 | Behavioural (sandbox) | SKIPPED |
What the source does
- Filesystem
- declared (14 observation(s))
- Network
- declared (5 observation(s))
- Shell
- declared (13 observation(s))
- Dependencies
- not all pinned
- Secrets in source
- found
Findings (25)
-----BEGIN RSA PRIVATE KEY-----
-----BEGIN PRIVATE KEY-----
-----BEGIN OPENSSH PRIVATE KEY-----
return yaml.load(content);
const frontMatter = yaml.load(frontMatterMatch[1]);
return yaml.load(content);
const frontMatter = yaml.load(frontMatterMatch[1]);
exec(`${editor} "${epicFile}"`, (error) => {exec(`${editor} "${issueFile}"`, (error) => {exec(`${editor} "${epicFile}"`, (error) => {exec(`${editor} "${issueFile}"`, (error) => {{ path: 'id_rsa', expected: CONFIG.permissions.privateKey, type: 'private key' },{ path: 'id_ed25519', expected: CONFIG.permissions.privateKey, type: 'private key' },{ path: 'id_ecdsa', expected: CONFIG.permissions.privateKey, type: 'private key' },{ path: 'id_rsa.pub', expected: CONFIG.permissions.publicKey, type: 'public key' },{ path: 'authorized_keys', expected: CONFIG.permissions.authorizedKeys, type: 'authorized keys' },DATABASE_URL: postgresql://postgres:postgres@postgres:5432/test
DATABASE_URL: postgresql://postgres:postgres@postgres:5432/test
IMPORTANT: Always include --label parameter with "task,epic:$ARGUMENTS"
IMPORTANT: Always include --label parameter with "task,epic:$ARGUMENTS"
console.log(`Token: stored in state file (do not share)`);
console.log(`Open ${baseUrl}/?token=<token> — token is in the state file: ${pidFile}`);console.log(`Could not open browser. The access URL (with token) is saved in: ${pidFile}`);console.log(`│ Token: ${this.padRight(hasToken, 22)} │`);console.log(`│ Token: ${this.padRight(hasToken, 22)} │`);Gates applied: critical_finding, no_behavioural_pass.
48b4be9853eefull audit observations/trust-audit/mcp-server/rafeekpro__claudeautopm.json · Report an issue / request a re-scanAudit history
Every audit this server has had. A grade with a past is a grade somebody is still checking.
| Date | Source | Verdict | Grade | Score | Change |
|---|---|---|---|---|---|
| 2026-10-12 | 48b4be9853ee | BLOCK | F | 30 | first audit |
Questions
What is the claude-autopm MCP server?
Autonomous Project Management Framework for Claude Code - Advanced AI-powered development automation
What tools does claude-autopm expose?
27 in total: 22 read-only, 5 that write, and 0 that can delete or overwrite. Every one is listed on this page with its risk.
Is claude-autopm safe to connect to an agent?
No — not without reading the findings first. The audit graded it F (30/100) and found 20 critical or high issues in the source. Each one is listed on this page with the file and line it is on.
What credentials does claude-autopm need?
It reads ANTHROPIC_API_KEY, AUTOPM_MASTER_PASSWORD, AZURE_DEVOPS_TOKEN, CONTEXT7_API_KEY, CUSTOM_KEY, GITHUB_TOKEN, OPENAI_API_KEY and TEST_API_KEY from the environment. Give it a token scoped to the least it needs — an agent that can be talked into calling a tool can be talked into calling it with your credentials.
How does claude-autopm run?
It speaks stdio, so it runs as a local process your client starts. It is published on npm as @claudeautopm/plugin-testing at 4.0.0.
How current is this page?
The grade is for one exact copy of the source (48b4be9853ee), read on 2026-10-12. The repository is watched and re-audited when it changes.