Good fit if
- Operators giving agents access to real tools
- Safety leads who need written, reviewable permission policy
- Agents enforcing a documented matrix
Safety & Security
# 2 of 27 · launch sequence
Agent Tool-Use Permission Matrix
A drop-in matrix that maps every tool an agent can call to a risk tier, an approval gate, and an audit requirement - enforceable by an orchestrator from one JSON file, deny-by-default.
v1.0.0-rc1
High risk
15-min quick start
Select a one-time tier below
View agent record
Built for: agent-platform-engineers · safety-leads · ai-agent-operators · orchestrator-agents ·
agent://products.json
200 OK
{
"id": "agent-tool-use-permission-matrix",
"version": "1.0.0-rc1",
"price_usd": { "lite":29, "std":79, "prem":129 },
"risk_level": "high",
"reversible": true,
"checkout": { "enabled": true },
"purchasable": true
}
See how it fits your agent stack
Tool call / action
requested by an agent
Agent Tool-Use Permission Matrix
TRL-5 method
Permission-Matrices/
Risk-Tiers/
Worked-Matrices/
Gated action
allowed or denied, logged
Human approval gate
Preview-before-apply
Rollback path
Discover
read the record
Evaluate
High risk · 6 systems
Install
new Permission-Matrices/
Preview
diff before apply
Approve
human gate
Audit & rollback
MXS-100 audit
Non-destructive by rule — new-folder install, never overwrite, human-approval gates for risky steps, rollback included.
agent-tool-use-permission-matrix/
Permission-Matrices/
Risk-Tiers/
Worked-Matrices/
README.md
AGENT-INSTRUCTIONS.md
Integration/ROLLBACK-GUIDE.md
Format & setup
FormatMarkdown + JSON file kit (plain text, tool-agnostic)
Skill levelIntermediate; beginner-usable via quick start
Quick start15 min
Full setup90 min
TRL-5 anchored risk ladder, machine-readable matrix contract + audit log schema, deny-by-default and allowlist/blocklist policies, agent authority tiers, exception lifecycle with mandatory expiry, misuse incident path, MXS-100 scorecard, 2 worked matrices + worked exception lifecycle, 16-doc integration pack, 5 upgrade modules, 10-doc differentiation folder
# Non-destructive lifecycle — nothing runs without preview + approval
$ agent read ./agent-tool-use-permission-matrix/manifest.json # metadata, risk, compatibility
$ agent install --new-folder ./agent-tool-use-permission-matrix # never overwrites existing files
$ agent apply --preview # preview-before-apply diff
$ agent verify ./agent-tool-use-permission-matrix --scorecard MXS-100 # Permission-Matrices contracts
⚠ gated steps require named human approval before execution
$ agent rollback ./agent-tool-use-permission-matrix # Delete the install folder (complete remo
✓ folder removed — nothing else touched
{
"required_permissions": ["write access to one new buyer-chosen folder"],
"integration_modes": ["standalone", "drop-in-upgrade", "advanced-integration", ],
"compatible_systems": [
"Claude (Projects, Skills, Claude Code)",
"ChatGPT (Custom GPTs, Projects)",
"any LLM chat assistant",
"MCP-based agent stacks",
],
"license_type": "single-operator",
"checkout": { "enabled": true, "status": "polar_active" }
}
The problem it solves
All-or-nothing tool access: no per-tool risk tiers, no approval mapping, no audit obligation on dangerous calls, no exception discipline.
Use cases
tier a new MCP server's tools before first use
scope sub-agent toolsets
run temporary-elevation exceptions with expiry
audit trails for dangerous calls
pre-map wallet/spend tools to human approval
answer a security review with one file
Safety & permissions
Non-destructive by rule
Plain-text kit; governs, never executes. Deny-by-default. T4 (spend/publish/delete/credential-adjacent) always requires named human approval; this floor cannot be lowered. Never handles credentials, keys, or wallets. Defensive use only.
Required
write access to one new buyer-chosen folder
Rollback
Delete the install folder (complete removal). Deeper integrations: Integrations/ROLLBACK-GUIDE.md.
For an evaluating agent
File kit mapping every agent-callable tool to an anchored risk tier (TRL-5), approval gate, and audit duty in one enforceable JSON matrix. Deny-by-default; T4 human floor; exception lifecycle with expiry. Extends P03 to tool level. Non-destructive install, delete-folder rollback, no credentials. Production: live and purchasable through verified Polar checkout; verified Polar checkout is active on this page.
Buyer decision
Is Agent Tool-Use Permission Matrix right for me?
Problem addressed: All-or-nothing tool access: no per-tool risk tiers, no approval mapping, no audit obligation on dangerous calls, no exception discipline.
Probably not a fit if
- Buyers expecting runtime enforcement software
- Stacks with no tool-using agents
- Anyone wanting an off switch they never have to configure
Included
Permission-Matrices/, Risk-Tiers/, Worked-Matrices/ folders with a machine-readable matrix contract and audit schema plus the standard architecture.
Not included
- No software, app, executable, or code that runs by itself
- No hosted service, account, login, or cloud dashboard
- No live integrations: Zapier/Make items are written blueprint descriptions, not built Zaps or scenarios
- No API access, API keys, credentials, or third-party subscriptions
| Tier | Price | Canonical depth | License |
|---|---|---|---|
| Standard | $79 USD | Full product: all templates, harnesses, SOPs, integration pack, and samples. | Single-Operator |
| Premium | $129 USD | Standard plus all 5 upgrade modules (Basic, Pro, Agent, Automation, Safety) and priority support. | Single-Operator |
Bundle membership
Compatibility and limits
Enforcement happens in YOUR runtime - this kit defines the policy, your stack applies it No code hooks or middleware included
Three one-time tiers — verified Polar checkout active
Lite
$29
Core templates & quick-start path. Best for testing the system manually first.
Buy Lite — one-timeStandardpopular
$79
Full product: all templates, harnesses, SOPs, integration pack, and samples.
Buy Standard — one-timePremium
$129
Standard plus all 5 upgrade modules (Basic, Pro, Agent, Automation, Safety) and priority support.
Buy Premium — one-timeEach tier uses its verified Polar checkout. See full pricing and bundles.
Related products
AI Agent Permission and Safety Gate Kit
Safety & Security
Agent Memory Governance Kit
Infrastructure & Governance
Agent Error Recovery Playbook
Reliability & Operations
In bundles: Wave 1 Agent Infrastructure Stack ·
Decision guides involving this product
- Agent Tool Permissions vs Safety GatesCompare ongoing tool authorization policy with action-specific approval and stop gates for higher-risk agent workflows.
- How to Choose an AI Agent Safety ToolkitSelect the right combination of permissions, defensive handling, audit evidence, safety cases, and deployment-readiness controls.
Related categories
- AI Agent Safety & GovernancePractical policy, evidence, permission, and defensive systems for reviewing how agents act and where they must stop.
- Multi-Agent SystemsRole, authority, handoff, recovery, and workflow standards for teams of agents and humans sharing work.
Available in canonical bundles
These bundle memberships come directly from the canonical bundle catalog. Compare the exact contents before purchase.