Audit whether a repository's environment variable configuration allows an agent to load all required .env variables from scratch. Use when evaluating a new repository's agent-readiness, when environment loading fails during agent setup, or as part of an initial harness assessment. Do not use when .env files already load correctly and the goal is to add a single new variable, or when creating env loading capability (use autonomy-env-creator).
Audit whether a repository's environment variable configuration allows an agent to load all required .env variables from scratch. Use when evaluating a new repository's agent-readiness, when environment loading fails during agent setup, or as part of an initial harness assessment. Do not use when .env files already load correctly and the goal is to add a single new variable, or when creating env loading capability (use autonomy-env-creator).
Environment Auditor
Evaluate whether an agent can configure and load all required environment variables from a cold start — with zero pre-existing .env files or manual setup.
Discipline
This auditor follows code-mint's shared auditor discipline:
Depth option — default standard; use quick for a sanity check on a recently audited repo or deep for legacy codebases, quarterly refreshes, or high-risk readiness claims. Heritage classification from meta-onboarding Phase 0 sets a sensible default.
Calibration — every report ends with a Calibration section naming the depth used, confidence level, what was not checked, and what would raise confidence. This is a required artifact, not a disclaimer.
This discipline is inspired by AWS AI-DLC's adaptive-depth and overconfidence-prevention ideas, but code-mint does not vendor or implement the full AI-DLC workflow.
Step 1: Discover Environment Requirements
Find all environment variable references in the codebase:
Search for process.env., os.environ, os.Getenv, ENV[, or equivalent patterns.
Check configuration files: scripts/env.sh, .env.example, .env.sample, docker-compose.yml, CI config.
What it connects to (database, API key, feature flag, etc.)
Whether it contains a secret (credentials, tokens, API keys)
Step 2: Evaluate Environment Loading Mechanism
Check how the project loads environment variables:
Primary setup exists: Is there a documented command or script (scripts/env.sh, make env, etc.) that produces local env config from the project's source of truth?
Secret source is documented: Is the secret system or provisioning flow clear (SSM, Vault, Doppler, 1Password, etc.)?
Setup covers all variables: Does the primary setup plus local overrides cover ALL required variables from Step 1?
Setup is documented: Does the README or AGENTS.md explain how to run it?
Prerequisites are documented: Required CLIs, auth steps, and local tools.
Loading mechanism documented: Is the tool/library used to load .env documented (dotenv, direnv, mise, etc.)?
Template fallback: If no primary setup exists, is there at least a .env.example or .env.sample template file?
Step 3: Test Cold Start Loading
Caution: Starting the application may have side effects (database creation, external API calls, email sends). Only attempt startup if you are confident it is safe in the current environment (e.g., local development with no production credentials). If unsure, evaluate readiness without starting the application and note that a live test was not performed.
Attempt to load the environment from scratch:
Simulate a cold start: If a documented bootstrap command exists, run it. If only a .env.example exists, note that the repo relies on a template-based flow and evaluate whether it is sufficient for local setup. Only copy the template to .env if that is clearly safe and expected for the project. If safe, attempt to start the application. If not safe, check whether the configuration files parse correctly without starting.
Identify failures: Which variables cause failures when missing or set to placeholder values?
Categorize blockers:
Hard blockers: Application crashes or refuses to start (database connection, required API keys)
Soft blockers: Application starts but features are degraded (analytics, email, optional integrations)
Non-blockers: Variables with sensible defaults that work in development
Step 4: Evaluate Secret Accessibility
Secrets are not hardcoded in source files, committed .env files, or documentation
Secret retrieval is documented: The process for obtaining each secret is described somewhere
Secret rotation is possible: Secrets can be rotated without code changes
Development secrets exist: There are development/test values for secrets that allow local development without production credentials
Ensure .gitignore ignores generated report contents while preserving the directories with their .gitkeep files.
Write the report to .agents/reports/autonomy-env-auditor-audit.md:
# Environment Audit Report
**Repository:** [name]
**Date:** [timestamp]
**Overall Status:** [Pass / Partial / Fail]
## Summary
- Total Variables Found: [N]
- Primary Setup: [Script / Command / Template Only / None]
- Secrets Identified: [N]
- Cold Start Result: [Success / Partial / Failure]
## Variable Inventory
| Variable | Required | Has Default | Secret | Documented | Status |
|---|---|---|---|---|---|
| DATABASE_URL | Yes | No | Yes | No | Missing from primary setup |
| ... | ... | ... | ... | ... | ... |
## Findings
### [Finding Title]
- **Severity:** [Critical / High / Medium / Low]
- **Current State:** [what exists]
- **Required State:** [what should exist]
- **Recommended Action:** [specific step]
- **Next Skill / Step:** [e.g., Run `autonomy-env-creator`]
## Cold Start Blockers
[List of variables that prevent application startup]
## Next Steps
Run `autonomy-env-creator` to remediate findings.
## Calibration
**Depth:** [quick / standard / deep]
**Confidence:** [High / Medium / Low] — [one-sentence reason]
**Not checked:**
- [specific area + reason]
**To raise confidence:**
- [specific next step]
After writing the report, update docs/onboarding-checklist.md and .agents/code-mint-status.json with the current smoke_path outcome status and date. Optionally update docs/skills-status.md if the repository keeps the compatibility view.
