القائمة
Secrets and authentication management. Ensures API keys, tokens, and credentials are NEVER exposed in plain text. Use when handling any credential, API key, or sensitive configuration.
التثبيت باستخدام Codex أو Claude انسخ هذا Prompt والصقه في Codex أو Claude أو مساعد آخر ليراجع صفحة Skill ويثبّتها لك.
استنادا إلى تصنيف SOC المهني
Intelligent routing between Bravo and Codex — decides when to delegate tasks to Codex vs handle internally
Use when CC asks to create a new agent, scaffold a new agent repo, or clone Bravo's architecture for a new domain (client agent, sibling agent, specialized agent). Generates a new AI-agent repo from templates, wires it into C_SUITE_ARCHITECTURE.md and brain/APP_REGISTRY.md, preserves the V5.6 outbound chokepoint, and gives the new agent a doctor command on day one.
Claims-based access control for multi-agent coordination. Defines what each agent can read, write, execute, and spawn. Enforces least-privilege: agents only access what they need. Use when: spawning agents, multi-agent tasks, security-sensitive operations. Skip when: single-agent inline work, trivial tasks.
Build and maintain product-grade agent infrastructure: onboarding diagnostics, runtime home, packaging, skill lifecycle, tool manifests, and agent scaffolds.
Spawn and coordinate Claude Code Agent Teams (experimental) — parallel subagents for complex multi-domain tasks
Creating algorithmic art using p5.js with seeded randomness and interactive parameter exploration. Use this when users request creating art using code, generative art, algorithmic art, flow fields, or particle systems. Create original algorithmic art rather than copying existing artists' work to avoid copyright violations.
| name | security-protocol |
| description | Secrets and authentication management. Ensures API keys, tokens, and credentials are NEVER exposed in plain text. Use when handling any credential, API key, or sensitive configuration. |
| triggers | ["secret","credential","API key","exposed","rotation","security","token","password","leak","gitguardian"] |
| tier | core |
| dependencies | [] |
Purpose: Ensures API keys, tokens, and database credentials are NEVER exposed in plain text files when different AI platforms interact with this workspace.
CLAUDE.md, ANTIGRAVITY.md, GEMINI.md) or any script. No api_key = "sk-..." lines anywhere..env.agents file. Never put them in code, never in commit messages..long_lived_token.txt, credentials.json, service_account.json should never exist inside a repo. If a tool needs such a file, put it outside the repo and reference it by path in .env.agents..env.agents or local shell env vars — never paste tokens into JSON configs..env.agents is gitignored everywhere. The repo root .gitignore in every agent must include .env + .env.* with only .env.agents.template whitelisted.--disallowedTools for Read/Grep/Glob/Bash/Edit/Write/WebFetch/…), a secret-stripped env, and a sandboxed cwd — not merely a read-only allowlist. A Read/Grep allowlist still reaches .env.agents by absolute path and exfiltrates it. Only operator-gated spawns (CC's Telegram id / an identity-gated /chat) get file tools. Reference: coordination_agent.js UNTRUSTED_DENY_TOOLS.Every agent ships with scripts/scan_secrets.py. Run it before shipping anything sensitive:
# Fast: scan the working tree (skips gitignored files)
python scripts/scan_secrets.py
# Thorough: scan every commit in every branch (catches leaked-once-then-deleted)
python scripts/scan_secrets.py --history
# Scan a sibling agent's repo
python scripts/scan_secrets.py --path ~/CMO-Agent
The scanner catches (non-exhaustive):
sk-ant-, sk-, AIza)ghp_, gho_, github_pat_)sbp_)sk_live_, sk_test_)AKIA, ASIA)EAA...) ← the 2026-04-24 CMO-Agent leak pattern*_token.txt, credentials.json, id_rsa, etc.)The leaked value is already public. Scrubbing git history on its own is insufficient — attackers may have already cloned. Revoke first.
/revokepip install git-filter-repo
git branch emergency-backup-before-scrub
git filter-repo --path <leaked-file> --invert-paths --force
git push origin --force --all
git push origin --force --tags
Add the leaked file's pattern to .gitignore, then commit + push.
If the leak affected anything client-facing (Stripe, client tokens, client data), notify CC immediately. Write a Reflexion entry in memory/MISTAKES.md documenting the root cause.
.gitignore of every new repo)# Secrets
.env
.env.*
!.env.agents.template
!.env.example
# Token/credential files
*.token
*_token.txt
*_token.json
*token*.txt
.long_lived_token*
credentials.json
service_account.json
# SSH / TLS
id_rsa
id_ed25519
*.pem
*.key
*.p12
*.pfx
# MCP configs (often carry API keys)
.claude/mcp.json
.vscode/mcp.json
templates/agent-scaffold/.gitignore ships with all of this — every forged agent inherits it.
2026-04-24 — CMO-Agent Facebook token leak
.long_lived_token.txt (183-char Facebook long-lived User Access Token) was committed in Maven's initial commit 3e6e83e on 2026-04-18. When CMO-Agent was flipped public later (to enable the cross-agent OASIS AI setup wizard clone flow), the token became world-readable. GitGuardian flagged it..env* was gitignored but not *.token / *token*.txt. The filename didn't trip the narrow gitignore.scripts/scan_secrets.py + hardened .gitignore patterns + scanner in bravo setup workflow + this skill updated..gitignore via templates/agent-scaffold/. Pre-push scans recommended before any public flip.When spawning subagents that need API access, pass secrets via environment variables or a shared .env.agents path reference — never paste the raw string into the prompt structure.
All outbound communications (emails, notifications, messages) referenced in this skill MUST be routed through
scripts/integrations/send_gateway.py. Directsmtplibor raw SMTP calls are architecturally prohibited (V5.6 chokepoint rule). Use:python scripts/integrations/send_gateway.py send --channel email --to <email> --subject "..." --body "..." --lead-id <uuid>See [[skills/send-gateway/SKILL.md]] for the full contract.