policyseed
Free, offline SOC 2 policy generator. policyseed renders 22 governance policy documentsfrom a single YAML intake file, using a small template engine and zero runtime dependencies.
npx policyseed init # writes intake.yaml: company, cloud, identity provider, MFA, team size...
npx policyseed build # renders the 22 policies into ./policies/
Prefer a browser? The same templates run at policyseed.io/generate: 23 questions, no signup, nothing uploaded.
- Free and Apache-2.0. Use it, fork it, modify it, ship it internally or to customers.
- Deterministic. The same
intake.yamlalways produces the same Markdown. No randomness. - No LLM. Templates are plain Markdown with a Mustache-style tag syntax, not model output.
- No signup. No network calls unless you collect evidence.
build,check,init,listandmcpmake no network calls and run entirely on your machine or CI runner.evidencecallsonly the cloud CLIs you already use (aws,gcloud,gh), with read-only commands, and onlywhen you run it.
The policies are a starting point that your company reviews and adopts, not a compliance guarantee. See What this is not.
What the output looks like
Every policy names your tools and people from the intake. An excerpt of the Access Control Policy,rendered by npx policyseed build for an example company (AWS, GitHub, Google Workspace, MFA on,11-50 people):
4.3 Google Workspace is the single source of workforce identity. Business applications areintegrated with it for single sign-on and, where supported, automated provisioning anddeprovisioning; applications that cannot federate are recorded in the exception register with acompensating review.
4.8 Access to GitHub is granted through organization membership enforced by single sign-onwith Google Workspace, repository permissions are assigned by team, and protected branchesrequire a review from someone other than the author before code is merged and deployed byGitHub Actions. Where headcount makes this impractical, the Security Owner documents thecompensating review.
All 22 policies share the same nine sections: purpose, scope, roles and responsibilities, numberedpolicy statements, procedures, exceptions, enforcement, review cadence and revision history. Thenumbered statements make it easy to point an auditor at the exact sentence behind a control.
Quick start
Requires Node.js 18 or later. No install step — run it with npx:
npx policyseed init
# edit intake.yaml with your company's details
npx policyseed build
npx policyseed check
initwrites a commentedintake.yamlin the current directory with every field, itsallowed values, and sensible defaults. It refuses to overwrite an existing file unless youpass--force.buildreadsintake.yaml, validates it, and writes all 22 policies to./policies/asNN-<slug>.mdfiles with a company-titled heading.checkreads./policies/andintake.yamland exits with status 1 if any policy's lastreviewed date is older than itsreview_cadenceallows. This is what the bundled GitHubAction runs on a schedule so an overdue policy shows up as a failing check.listprints the 22 bundled templates.evidencecollects read-only SOC 2 evidence from AWS, Google Cloud and GitHub with the CLIsyou already have signed in. See Collecting audit evidence.--helpand--versiondo what you'd expect.
If you'd rather install it as a dependency:
npm install --save-dev policyseed
npx policyseed init
Template syntax reference
Templates are plain Markdown files with YAML front matter and a small set of tags, implementedin lib/render.js:
| Tag | Behavior |
|---|---|
{{key}} |
Prints the value of key from the intake context. Unknown keys render as an empty string. |
{{#if key}} … {{/if}} |
Renders the block only when key is truthy. |
{{#unless key}} … {{/unless}} |
Renders the block only when key is falsy. |
{{#each list}} … {{/each}} |
Repeats the block once per item in list; {{this}} is the current item. |
Truthiness matches JavaScript, plus empty arrays count as falsy: false, "", 0, NaN,null, undefined and [] are falsy; everything else is truthy.
A block tag that stands alone on its own line (only whitespace before and after it) is removedtogether with its line, so conditional paragraphs and loops don't leave stray blank linesbehind. Three or more consecutive blank lines collapse to two.
The context available to every template is built by toTemplateContext() inlib/context.js from your intake.yaml — company name, cloud providers,identity provider, MFA status, data types handled, and so on, plus derived booleans likehas_mdm, has_vendors, scope_availability and small_team (headcount 1-10) that templates branch on.
The 22 policies
| # | Title |
|---|---|
| 1 | Information Security Policy |
| 2 | Acceptable Use Policy |
| 3 | Access Control Policy |
| 4 | Authentication and Password Policy |
| 5 | Asset Management Policy |
| 6 | Data Classification and Handling Policy |
| 7 | Data Retention and Disposal Policy |
| 8 | Encryption and Key Management Policy |
| 9 | Change Management Policy |
| 10 | Secure Software Development Policy |
| 11 | Vulnerability and Patch Management Policy |
| 12 | Logging and Monitoring Policy |
| 13 | Incident Response Policy |
| 14 | Business Continuity and Disaster Recovery Policy |
| 15 | Backup and Recovery Policy |
| 16 | Vendor and Third-Party Risk Management Policy |
| 17 | Risk Assessment and Management Policy |
| 18 | Human Resources Security Policy |
| 19 | Endpoint and Workstation Security Policy |
| 20 | Network and Infrastructure Security Policy |
| 21 | Physical and Remote Work Security Policy |
| 22 | Privacy and Data Protection Policy |
Each one has nine standard sections: Purpose, Scope, Roles and Responsibilities, PolicyStatements, Procedures, Exceptions, Enforcement, Review Cadence, and Revision History.
Keeping policies current in CI
Use the GitHub Action in a repo that keeps intake.yaml and policies/ under version control:
- uses: actions/checkout@v4
- uses: odedmoshe/policyseed@v1
with:
command: check
It fails the check when a policy is overdue for review, so a stale policy shows up as a red Xinstead of quietly rotting. .github/workflows/policy-review.ymlis a complete weekly workflow to copy; inputs and outputs are indocs/github-action.md.
Use it from an AI assistant (MCP)
policyseed mcp starts a Model Context Protocol server, soClaude, Cursor and other MCP clients can list the policies, ask you the intake questions andrender any policy in the conversation. Nothing leaves your machine except what you type to theassistant.
claude mcp add policyseed -- npx -y policyseed mcp
Claude Desktop and Cursor configuration, the tool list and an example conversation are indocs/mcp.md.
Agent skill
skills/soc2-policies/ is an agent skill (a SKILL.md plus aninterview guide) that tells Claude Code and other skill-aware agents when to use Policyseed andhow: interview for the intake, render with the CLI or the MCP tools, then report what to review.To install it for Claude Code, copy the folder into ~/.claude/skills/ (or .claude/skills/ in aproject).
Collecting audit evidence
Auditors ask for proof that controls are configured: branch protection, MFA, encryption at rest,logging, who has admin access. policyseed evidence gathers that with the aws, gcloud and ghCLIs you already have installed and signed in, and saves it as dated JSON files plus an index:
npx policyseed evidence --providers aws,gcp,github \
--repo acme/app --org acme --project acme-prod
evidence/2026-10-23/
INDEX.md criterion -> file -> what it shows -> auditor question
CC6.1-aws-iam-account-summary.json
CC6.1-aws-iam-credential-report.json
CC8.1-github-branch-protection.json
...
How it behaves:
- Read-only. Every command is a Describe, Get or List call (
gh apiGET requests, no requestfields). The one exception in name only isaws iam generate-credential-report, which asks IAMto build the credential report; IAM classifies it as a Read action and it changes no settings.--dry-runprints the exact commands without running anything;--listprints the collectors. - No shell. Each command runs with Node's
execFileand a fixed argument array.--repo,--org,--project, bucket names and trail ARNs are passed as separate arguments, neverinterpolated into a shell string. Pagers and interactive prompts are turned off. - Each file wraps the CLI's JSON output with the command that produced it, the time, the CLIversion and the exit code. Files are named
<criterion>-<collector-id>.jsonafter thecollector's first SOC 2 criterion; INDEX.md lists every criterion a file supports. - Redaction. Before anything is written or printed, values under secret-looking keys(
SessionToken,SecretAccessKey,password,token, ...) are replaced, and so are stringsthat look like credentials: GitHub tokens, AWS access key IDs and session tokens, Google API keysand OAuth tokens, JWTs, PEM private keys andkey=valuesecrets. Secrets are never requested inthe first place. Account IDs are kept, because auditors usually need them; pass--redact-account-idsto truncate 12-digit IDs to their last four digits. - Failures don't stop the run. A missing CLI, missing credentials or a permission error iswritten to INDEX.md as
not collected: <reason>and the run moves on. The exit code is 1 onlyif nothing at all was collected. - Re-runs. A run replaces the files its providers wrote earlier the same day and rewritesINDEX.md. Other days' folders, other providers' files and any files you add yourself (such asscreenshots) are left alone.
- Scope. Regional AWS checks (EBS default encryption, RDS, CloudTrail) use the CLI'sconfigured region (
AWS_REGIONor the profile's region). GitHub checks cover one repository andone organization; Google Cloud checks cover one project (--project, or the gcloud default).S3 per-bucket checks stop after--max-bucketsbuckets (default 50). On Windows,gcloudis a.cmdscript that Node can't start without a shell, so the Google Cloud collectors report it asnot collected there.
What it collects
| Collector | Criteria | Command |
|---|---|---|
github-repo-settings |
CC7.1 | gh api repos/{repo}: default branch, secret scanning and push protection |
github-branch-protection |
CC8.1 | gh api repos/{repo}/branches/{default}/protection |
github-branch-rules |
CC8.1 | gh api --paginate repos/{repo}/rules/branches/{default} (ruleset rules) |
github-dependabot-alerts |
CC7.1 | gh api repos/{repo}/vulnerability-alerts |
github-org-2fa |
CC6.1 | gh api orgs/{org}: two_factor_requirement_enabled |
github-org-admins |
CC6.3 | gh api --paginate orgs/{org}/members?role=admin |
aws-iam-account-summary |
CC6.1 | aws iam get-account-summary: root MFA, root access keys |
aws-iam-credential-report |
CC6.1, CC6.2 | aws iam generate-credential-report, then get-credential-report, decoded and reduced to the MFA, password and access-key age columns |
aws-iam-password-policy |
CC6.1 | aws iam get-account-password-policy |
aws-cloudtrail |
CC7.2 | aws cloudtrail describe-trails, then get-trail-status per trail |
aws-ebs-default-encryption |
CC6.1 | aws ec2 get-ebs-encryption-by-default |
aws-s3-buckets |
CC6.1, CC6.6 | aws s3api list-buckets, then get-bucket-encryption and get-public-access-block per bucket |
aws-rds-instances |
CC6.1, A1.2 | aws rds describe-db-instances: StorageEncrypted, BackupRetentionPeriod and related fields |
gcp-project-iam-policy |
CC6.3 | gcloud projects get-iam-policy {project} |
gcp-logging-sinks |
CC7.2 | gcloud logging sinks list |
gcp-sql-instances |
A1.2, CC6.1 | gcloud sql instances list (backup configuration) |
gcp-compute-disks |
CC6.1 | gcloud compute disks list (encryption keys) |
gcp-organizations |
CC6.1 | gcloud organizations list |
All aws commands add --output json and all gcloud commands add --format=json.
Permissions it needs
Give the identity you run it with read access only.
AWS. This policy covers every AWS collector and nothing else:
{
"Version": "2012-10-17",
"Statement": [
{
"Sid": "PolicyseedEvidenceReadOnly",
"Effect": "Allow",
"Action": [
"iam:GetAccountSummary",
"iam:GenerateCredentialReport",
"iam:GetCredentialReport",
"iam:GetAccountPasswordPolicy",
"cloudtrail:DescribeTrails",
"cloudtrail:GetTrailStatus",
"ec2:GetEbsEncryptionByDefault",
"s3:ListAllMyBuckets",
"s3:GetEncryptionConfiguration",
"s3:GetBucketPublicAccessBlock",
"rds:DescribeDBInstances"
],
"Resource": "*"
}
]
}
Google Cloud. roles/viewer on the project covers the project collectors, androles/resourcemanager.organizationViewer on the organization covers organizations list.For narrower access, grant roles/iam.securityReviewer (IAM policy), roles/logging.viewer(sinks), roles/cloudsql.viewer (SQL instances) and roles/compute.viewer (disks) on the project.
GitHub. gh uses whatever it is signed in with (gh auth login, or GH_TOKEN). Repositoryadmin read access is needed for classic branch protection, the security_and_analysis block andthe Dependabot alerts setting; the ruleset rules need only read access. For the organization,GitHub returns two_factor_requirement_enabled only to organization owners, and lists privatemembers only to members. A fine-grained token with repository Administration: read andMetadata: read and organization Members: read covers the rest. A collector the token can'tread is recorded as not collected, never as a pass.
To run the GitHub collectors on a schedule and keep the results as a workflow artifact, seedocs/evidence.md.
What it does not do
It saves configuration as it is on the day you run it. It doesn't judge whether thatconfiguration meets your policies, it doesn't cover every control an auditor will test, and itdoesn't replace the auditor's own testing: they will choose their own samples, may ask you tore-run commands while they watch, and will test whether the controls operated over the wholeaudit period.
Relationship to the hosted generator and Audit Kit
This package is the same rendering engine that powers the hosted generator atpolicyseed.io, packaged for people who'd rather run itlocally, script it into CI, or read exactly how it works; the hosted site also offers a paidAudit Kit that uses Claude to tailor the Policy Statements and Procedures sections to yourspecific stack and adds export formats — this CLI does not include that tailoring step.
What this is not
These are governance document templates, not legal advice, and not a compliance guarantee. A SOC 2report is an independent CPA firm's opinion, and it depends on whether you actually operate the controls these policies describe — access reviews happening onschedule, incidents actually being logged and triaged, backups actually being tested — not juston having the documents. Have a qualified person (counsel, an auditor, or an experiencedcompliance practitioner) review the output before you rely on it externally.
Contributing
Issues and pull requests are welcome atgithub.com/odedmoshe/policyseed. If you're proposinga change to a policy's wording, prefer editing the template's prose directly and running thetest suite (node --test test/) so the renderer parity and CLI smoke tests still pass. Ifyou're adding a new intake field, update lib/context.js (DEFAULT_INTAKE,toTemplateContext, validateIntake) together, since the three are meant to stay in lockstep.
License
Apache License 2.0 — see LICENSE. Copyright the policyseed contributors.