axiom-security

What the source does

• FilesystemNone observedNo reads or writes in the source
• NetworkNone observedNo outbound calls in the source
• ShellNone observedNo commands run in the source
• DependenciesPinnedNothing floats
• SecretsNone foundNo keys in source

Five audit layers

Gates applied: instruction override, no behavioural pass.

- Background access pitfalls, Mac keychain differences

→ Read: `skills/keychain.md`

→ Read: `skills/keychain-diag.md`

→ Read: `skills/keychain-ref.md`

- [ ] AASA changes propagated (or developer mode enabled for testing)

Instructions embedded in *extra context* given to the model — the initial context or any tool result — with the intent to redirect control flow. A calendar event titled "Ignore previous instructions a

**Pressure**: "Two engineers spent a day trying to jailbreak it and couldn't. That's our safety testing."

| "codesign wants to access key" dialog | Keychain access not granted to codesign |

Audit 0.4.1 · source sha 363465e88c85 · 31 Oct 2026Report an issue or request a re-scan

• Publishercharleswiltgen GitHub userOwner of the repository this skill is read from
• Source foldercharleswiltgen/axiom/axiom-codex/skills/axiom-securityThe exact folder we audited
• LicenceMITAs the repository states it
• Commit pinned for this verdict363465e88c85A new audit runs when the source changes
• Repository statusNot archivedPushed today
• Sandboxed runNot runSkills are read, not executed; layer L4 is skipped

Its description: Use when storing credentials securely, encrypting data, implementing passkeys, securing AI/agentic features against prompt injection, code signing, or managing certificates and provisioning profiles.

---
name: axiom-security
description: Use when storing credentials securely, encrypting data, implementing passkeys, securing AI/agentic features against prompt injection, code signing, or managing certificates and provisioning profiles.
license: MIT
---

# Security & Credentials

**You MUST use this skill for ANY keychain, encryption, passkey, app integrity, agentic/AI feature security, file protection, or code signing work.**

<!-- AXIOM_AUDITOR_INLINE_BEGIN — rewritten for Codex by scripts/build-codex.ts; do not hand-edit -->
> **Auditors are skills here.** Where this router says "Launch `some-auditor` agent", invoke the
> matching Codex skill instead — same procedure, no Claude Code agent required.
>
> Available: `axiom-scan-security-privacy`.
>
> The ones that shell out — builds, tests, simulators, crash symbolication — need shell access to run.
<!-- AXIOM_AUDITOR_INLINE_END -->

## Quick Reference

| Symptom / Task | Reference |
|----------------|-----------|
| Store tokens, passwords, API keys securely | See `skills/keychain.md` |
| Choose kSecAttrAccessible level, biometric protection | See `skills/keychain.md` |
| SecItem function signatures, attribute constants | See `skills/keychain-ref.md` |
| errSecDuplicateItem, errSecItemNotFound, errSecInteractionNotAllowed | See `skills/keychain-diag.md` |
| Encrypt data, sign payloads, key management | See `skills/cryptokit.md` |
| Hash functions, HMAC, AES-GCM, ChaChaPoly, ECDSA, EdDSA, key agreement | See `skills/cryptokit-ref.md` |
| Passkey sign-in, WebAuthn, ASAuthorizationController | See `skills/passkeys.md` |
| One-time code AutoFill for credential providers `OS27` | See `skills/passkeys.md` (Delivered Verification Codes) |
| App integrity verification, DCAppAttestService, fraud metric | See `skills/app-attest.md` |
| Prompt injection, securing AI agents / agentic features, tool confirmation | See `skills/agentic-security.md` |
| NSFileProtection levels, data protection at rest | See `skills/file-protection-ref.md` |
| SensitiveContentAnalysis verdict must never leave the device (no analytics/moderation queue/synced cache; license §3.3.3) | See `axiom-vision` (skills/vision-ref.md) |
| Certificate management, provisioning profiles, CI/CD signing | See `skills/code-signing.md` |
| Certificate not found, profile mismatch, entitlement errors | See `skills/code-signing-diag.md` |
| Certificate CLI, profile inspection, entitlement extraction | See `skills/code-signing-ref.md` |
| Apple Pay payment certs / pass type certs / Tap to Pay entitlement | See `axiom-payments` suite |

## Decision Tree

```dot
digraph security {
    start [label="Security task" shape=ellipse];
    what [label="What do you need?" shape=diamond];

    start -> what;
    what -> "skills/keychain.md" [label="store/retrieve\ncredentials, tokens,\nsecrets"];
    what -> "skills/keychain-ref.md" [label="SecItem API syntax,\nattribute constants,\naccess levels"];
    what -> "skills/keychain-diag.md" [label="keychain errors\n(errSec codes)"];
    what -> "skills/cryptokit.md" [label="encrypt data,\nsign payloads,\nSecure Enclave keys"];
    what -> "skills/cryptokit-ref.md" [label="CryptoKit API\n(AES, ECDSA, HPKE,\npost-quantum)"];
    what -> "skills/passkeys.md" [label="passkey sign-in,\nreplace passwords"];
    what -> "skills/app-attest.md" [label="app integrity,\nfraud prevention"];
    what -> "skills/agentic-security.md" [label="AI agent security,\nprompt injection"];
    what -> "skills/file-protection-ref.md" [label="file encryption,\nNSFileProtection"];
    what -> "skills/code-signing.md" [label="set up signing,\nprofiles, CI/CD"];
    what -> "skills/code-signing-diag.md" [label="signing errors,\nupload rejections"];
    what -> "skills/code-signing-ref.md" [label="CLI commands,\nprofile inspection"];
}
```

1. Store tokens, passwords, API keys securely? → `skills/keychain.md`
1a. Need SecItem function signatures, attribute constants? → `skills/keychain-ref.md`
1b. Keychain errors (errSecDuplicateItem, errSecItemNotFound)? → `skills/keychain-diag.md`
2. Encrypt data, sign payloads, manage keys? → `skills/cryptokit.md`
2a. Need CryptoKit API details (AES-GCM, ECDSA, HPKE, post-quantum)? → `skills/cryptokit-ref.md`
3. Implement passkey sign-in, replace passwords? → `skills/passkeys.md`
4. Verify app integrity, prevent fraud? → `skills/app-attest.md`
5. Securing an agentic/AI feature (prompt injection, tool confirmation, lock-screen intents)? → `skills/agentic-security.md`
6. File encryption at rest, NSFileProtection levels? → `skills/file-protection-ref.md`
7. Set up code signing, manage certificates, CI/CD? → `skills/code-signing.md`
7a. Code signing error troubleshooting? → `skills/code-signing-diag.md`
7b. Certificate CLI commands, profile inspection? → `skills/code-signing-ref.md`
8. Build/upload failures after signing? → See axiom-build
9. App Store submission prep? → `/skill axiom-shipping`
10. Privacy manifests, tracking transparency? → See axiom-integration
11. Data persistence (SwiftData, Core Data, storage strategy)? → `/skill axiom-data`
12. TLS configuration, certificate pinning for network requests? → `/skill axiom-networking`
13. Want automated security scan? → security-privacy-scanner (Agent)

## Conflict Resolution

**security vs axiom-build**: When build fails with signing errors:
- Code signing errors (certificate, profile, entitlement) → **use security**
- Environment issues (Xcode version, simulator, Derived Data) → **use axiom-build**
- If unsure, check the error message: `CODESIGN`, `ITMS-90xxx`, `errSec` → **security**

**security vs shipping**: When preparing for App Store:
- Privacy manifests, submission checklists, rejections → **use shipping**
- Code signing for distribution, certificate management → **use security**

**security vs axiom-data**: When storing sensitive data:
- Tokens, passwords, API keys → **use security** (keychain)
- User preferences, non-sensitive settings → **use axiom-data** (UserDefaults/SwiftData)
- File en

Excerpt, first 6,000 characters of SKILL.md, verbatim from the audited commit. It is text from someone else’s repository, quoted so you can judge it before your agent reads it.

1. 2026-10-31
   ✕BLOCK D 69 / 100

   first audit · source 363465e88c85

Questions about axiom-security

• github.com/charleswiltgen/axiom/tree/main/axiom-codex/skills/axi — the audited source at commit 363465e88c85 and every finding aboveObserved · 31 Oct 2026

Spotted something wrong? Report a correction · Methodology