pydanticai-tealtiger
Deterministic governance guard for Pydantic AI agents — policy enforcement, PII detection, cost tracking, kill switch, and structured audit evidence.
No LLM in the governance path. All policy evaluation is deterministic, adding <2ms latency per call.
Installation
pip install pydanticai-tealtiger
Quick Start
from pydantic_ai import Agent
from pydanticai_tealtiger import TealTigerGuard
# Zero-config: observe, track cost, detect PII, allow all
guard = TealTigerGuard()
agent = Agent('openai:gpt-4', deps_type=TealTigerGuard)
@agent.tool
async def search(ctx, query: str) -> str:
# Governance check before execution
ctx.deps.evaluate(tool="search", args={"query": query})
result = perform_search(query)
# Record cost after execution
ctx.deps.post_call(tool_name="search", result=result)
return result
# Run with governance
result = await agent.run("Find AI safety papers", deps=guard)
# Inspect governance data
print(guard.cumulative_cost)
print(guard.audit_trail)
print(guard.summary)
Features
| Feature | Description |
|---|---|
| Zero-config mode | Observe all tool calls, track cost, detect PII — no setup required |
| Policy mode | Evaluate TealEngine policies, block on DENY in ENFORCE mode |
| PII detection | Detect email, SSN, credit card, phone, IP in tool arguments |
| Cost tracking | Per-tool, per-session cost accumulation |
| Budget limits | Set max spend per session — deny when exceeded |
| Tool allowlist | Restrict which tools can be invoked |
| Kill switch | freeze() / unfreeze() to halt all tool execution |
| Audit trail | Structured entries with UUID v4 correlation IDs |
| TEEC namespace | teec.pydanticai evidence contract fields |
Governance Modes
| Mode | Behavior |
|---|---|
OBSERVE |
Allow all, log decisions, track cost (default) |
MONITOR |
Allow all, log denials as warnings |
ENFORCE |
Block denied actions, raise GovernanceDenyError |
Usage Patterns
Policy Mode with TealEngine
from tealtiger import TealEngine
from pydanticai_tealtiger import TealTigerGuard
engine = TealEngine(policies=[
{"type": "cost_limit", "max_per_session": 5.00},
{"type": "pii_block", "action": "DENY"},
])
guard = TealTigerGuard(engine=engine, mode="ENFORCE")
Tool Allowlist
guard = TealTigerGuard(
mode="ENFORCE",
tool_allowlist=["search", "compute", "read_file"],
)
Budget Limit
guard = TealTigerGuard(
mode="ENFORCE",
budget_limit=1.00, # $1.00 max per session
)
Kill Switch
guard = TealTigerGuard(mode="ENFORCE")
# Emergency stop — blocks all subsequent calls
guard.freeze()
# Resume operations
guard.unfreeze()
pre_call / post_call Pattern
@agent.tool
async def my_tool(ctx, arg: str) -> str:
# pre_call is an alias for evaluate
ctx.deps.pre_call(tool_name="my_tool", args={"arg": arg})
result = do_work(arg)
# Record actual cost
ctx.deps.post_call(
tool_name="my_tool",
result=result,
token_usage={"total_tokens": 500},
)
return result
API Reference
TealTigerGuard
| Parameter | Type | Default | Description |
|---|---|---|---|
engine |
TealEngine | None |
None |
Policy engine for evaluation |
mode |
str |
"OBSERVE" |
Governance mode |
cost_per_1k_tokens |
float |
0.002 |
Cost estimation rate |
session_id |
str | None |
auto-generated | Session identifier |
tool_allowlist |
list[str] | None |
None |
Allowed tool names |
budget_limit |
float | None |
None |
Max cost per session (USD) |
Methods
| Method | Description |
|---|---|
evaluate(tool, args, agent_id) |
Evaluate governance before tool execution |
pre_call(tool_name, args) |
Alias for evaluate() |
post_call(tool_name, result, token_usage) |
Record cost after tool execution |
freeze() |
Activate kill switch |
unfreeze() |
Deactivate kill switch |
Properties
| Property | Type | Description |
|---|---|---|
audit_trail |
list[AuditEntry] |
Full governance audit trail |
summary |
dict[str, ToolSummary] |
Cost/call counts per tool |
cumulative_cost |
float |
Total session cost |
session_id |
str |
Session identifier |
Development
cd packages/pydanticai-tealtiger
pip install -e ".[dev]"
pytest
License
Apache-2.0 — see LICENSE.
Metadata
Release files for pydanticai-tealtiger 0.1.0
For a detailed explanation of source distributions (sdists) and built distributions (wheels), please see the package formats documentation.
Source distribution (sdist)
| File | Size | Uploaded | Provenance | Actions |
|---|---|---|---|---|
| pydanticai_tealtiger-0.1.0.tar.gz | 20.1 kB | Without provenance | Details for pydanticai_tealtiger-0.1.0.tar.gz |
Built distribution (wheel)
| File | Interpreter | ABI | Platform | Actions Reset |
|---|---|---|---|---|
| pydanticai_tealtiger-0.1.0-py3-none-any.whl | Python 3 | none | any | Details for pydanticai_tealtiger-0.1.0-py3-none-any.whl |
Total release size: 34.4 kB
Release files / pydanticai_tealtiger-0.1.0.tar.gz
| Download URL | pydanticai_tealtiger-0.1.0.tar.gz |
|---|---|
| Size | 20.1 kB |
| Tags | Source |
|
SHA-256 checksum How to use checksums |
bdf24d9912b431c2bcd98ab2d8198ca828fd09714e1607a08a4c744037b22eda
|
|
BLAKE2b-256 checksum How to use checksums |
4a8fd1aa87f0248637dff447a0ab18eca3307b6ebb02269009e1dce351a7f42f
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
No |
| Uploaded via |
twine/6.2.0 CPython/3.14.2
|
Release files / pydanticai_tealtiger-0.1.0-py3-none-any.whl
| Download URL | pydanticai_tealtiger-0.1.0-py3-none-any.whl |
|---|---|
| Size | 14.3 kB |
| Tags | Python 3 |
|
SHA-256 checksum How to use checksums |
df0f768f18e3f5a70b0a39ce3f39026eb655ed5106de2b1fd8f5d8b0a372cefb
|
|
BLAKE2b-256 checksum How to use checksums |
f547689d7e2d63930c838f950fcea2c184ebb8e85c64f4d607e9e9705c616184
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
No |
| Uploaded via |
twine/6.2.0 CPython/3.14.2
|
Security for pydanticai-tealtiger 0.1.0
Provenance
Provenance describes where a file came from. On PyPI, provenance is shared via attestations, which provide a verifiable record of the build or publishing details. View details, limitations and caveats.
No provenance found
None of the files in this release were uploaded with provenance metadata.
A note on package security
PyPI is a repository for distributing packages, not a service for auditing them. While third-party security researchers and our staff do their best to detect and remove malicious packages, PyPI cannot guarantee the quality, security or suitability of any given package.
To learn more about protecting yourself from malicious actors, visit our package security documentation.