Skip to content

Your AI agent just burned $200. AgentGuard stops it at $5. Runtime cost guardrails for AI agents — budget enforcement, loop detection, kill switch. Zero dependencies, MIT licensed.

License

Notifications You must be signed in to change notification settings

bmdhodl/agent47

AgentGuard

Your AI agent just burned $200 in one run. AgentGuard would have stopped it at $5.

Set a dollar budget. Get warnings at 80%. Kill the agent when it exceeds the limit. Zero dependencies, works with any framework.

PyPI Downloads Python CI Coverage License: MIT OpenSSF Scorecard GitHub stars

pip install agentguard47

Try it in 60 seconds

No API keys. No config. Just run it:

pip install agentguard47 && python examples/try_it_now.py
Simulating agent making LLM calls with a $1.00 budget...

  Call 1: $0.12 spent
  Call 2: $0.24 spent
  ...
  WARNING: Budget warning: cost 84% of limit reached (threshold: 80%)
  Call 7: $0.84 spent
  Call 8: $0.96 spent

  STOPPED at call 9: Cost budget exceeded: $1.08 > $1.00
  Total: $1.08 | 9 calls

  Without AgentGuard, this agent would have kept spending.

Open In Colab

Quickstart: Stop Runaway Costs in 4 Lines

from agentguard import Tracer, BudgetGuard, patch_openai

tracer = Tracer(guards=[BudgetGuard(max_cost_usd=5.00, warn_at_pct=0.8)])
patch_openai(tracer)  # auto-tracks every OpenAI call

# Use OpenAI normally — AgentGuard tracks cost and kills the agent at $5

That's it. Every ChatCompletion call is tracked. When accumulated cost hits $4 (80%), your warning fires. At $5, BudgetExceeded is raised and the agent stops.

No config files. No dashboard required. No dependencies.

The Problem

AI agents are expensive and unpredictable:

  • Cost overruns average 340% on autonomous agent tasks (source)
  • A single stuck loop can burn through your entire OpenAI budget in minutes
  • Existing tools (LangSmith, Langfuse, Portkey) show you the damage after it happens

AgentGuard is the only tool that kills agents mid-run when they exceed spend limits.

AgentGuard LangSmith Langfuse Portkey
Hard budget enforcement Yes No No No
Kill agent mid-run Yes No No No
Loop detection Yes No No No
Cost tracking Yes Yes Yes Yes
Zero dependencies Yes No No No
Self-hosted option Yes No Yes No
Price Free (MIT) $2.50/1k traces $59/mo $49/mo

Guards

Guards are runtime checks that raise exceptions when limits are hit. The agent stops immediately.

Guard What it stops Example
BudgetGuard Dollar/token/call overruns BudgetGuard(max_cost_usd=5.00)
LoopGuard Exact repeated tool calls LoopGuard(max_repeats=3)
FuzzyLoopGuard Similar tool calls, A-B-A-B patterns FuzzyLoopGuard(max_tool_repeats=5)
TimeoutGuard Wall-clock time limits TimeoutGuard(max_seconds=300)
RateLimitGuard Calls-per-minute throttling RateLimitGuard(max_calls_per_minute=60)
from agentguard import BudgetGuard, BudgetExceeded

budget = BudgetGuard(
    max_cost_usd=10.00,
    warn_at_pct=0.8,
    on_warning=lambda msg: print(f"WARNING: {msg}"),
)

# In your agent loop:
budget.consume(tokens=1500, calls=1, cost_usd=0.03)
# At 80% → warning callback fires
# At 100% → BudgetExceeded raised, agent stops

Integrations

LangChain

pip install agentguard47[langchain]
from agentguard import Tracer, BudgetGuard
from agentguard.integrations.langchain import AgentGuardCallbackHandler

tracer = Tracer(guards=[BudgetGuard(max_cost_usd=5.00)])
handler = AgentGuardCallbackHandler(
    tracer=tracer,
    budget_guard=BudgetGuard(max_cost_usd=5.00),
)

# Pass to any LangChain component
llm = ChatOpenAI(callbacks=[handler])

LangGraph

pip install agentguard47[langgraph]
from agentguard.integrations.langgraph import guarded_node

@guarded_node(tracer=tracer, budget_guard=BudgetGuard(max_cost_usd=5.00))
def research_node(state):
    return {"messages": state["messages"] + [result]}

CrewAI

pip install agentguard47[crewai]
from agentguard.integrations.crewai import AgentGuardCrewHandler

handler = AgentGuardCrewHandler(
    tracer=tracer,
    budget_guard=BudgetGuard(max_cost_usd=5.00),
)

agent = Agent(role="researcher", step_callback=handler.step_callback)

OpenAI / Anthropic Auto-Instrumentation

from agentguard import Tracer, BudgetGuard, patch_openai, patch_anthropic

tracer = Tracer(guards=[BudgetGuard(max_cost_usd=5.00)])
patch_openai(tracer)      # auto-tracks all ChatCompletion calls
patch_anthropic(tracer)   # auto-tracks all Messages calls

Cost Tracking

Built-in pricing for OpenAI, Anthropic, Google, Mistral, and Meta models. Updated monthly.

from agentguard import estimate_cost

# Single call estimate
cost = estimate_cost("gpt-4o", input_tokens=1000, output_tokens=500)
# → $0.00625

# Track across a trace — cost is auto-accumulated per span
with tracer.trace("agent.run") as span:
    span.cost.add("gpt-4o", input_tokens=1200, output_tokens=450)
    span.cost.add("claude-sonnet-4-5-20250929", input_tokens=800, output_tokens=300)
    # cost_usd included in trace end event

Tracing

Full structured tracing with zero dependencies — JSONL output, spans, events, and cost data.

from agentguard import Tracer, JsonlFileSink, BudgetGuard

tracer = Tracer(
    sink=JsonlFileSink("traces.jsonl"),
    guards=[BudgetGuard(max_cost_usd=5.00)],
)

with tracer.trace("agent.run") as span:
    span.event("reasoning", data={"thought": "search docs"})
    with span.span("tool.search", data={"query": "quantum computing"}):
        pass  # your tool logic
    span.cost.add("gpt-4o", input_tokens=1200, output_tokens=450)
$ agentguard report traces.jsonl

AgentGuard report
  Total events: 9
  Spans: 6  Events: 3
  Estimated cost: $0.01

Evaluation

Assert properties of your traces in tests or CI.

from agentguard import EvalSuite

result = (
    EvalSuite("traces.jsonl")
    .assert_no_loops()
    .assert_budget_under(tokens=50_000)
    .assert_completes_within(seconds=30)
    .assert_no_errors()
    .run()
)
agentguard eval traces.jsonl --ci   # exits non-zero on failure

CI Cost Gates

Fail your CI pipeline if an agent run exceeds a cost budget. No competitor offers this.

# .github/workflows/cost-gate.yml (simplified)
- name: Run agent with budget guard
  run: |
    python3 -c "
    from agentguard import Tracer, BudgetGuard, JsonlFileSink
    tracer = Tracer(
        sink=JsonlFileSink('ci_traces.jsonl'),
        guards=[BudgetGuard(max_cost_usd=5.00)],
    )
    # ... your agent run here ...
    "

- name: Evaluate traces
  uses: bmdhodl/agent47/.github/actions/agentguard-eval@main
  with:
    trace-file: ci_traces.jsonl
    assertions: "no_errors,max_cost:5.00"

Full workflow: docs/ci/cost-gate-workflow.yml

Async Support

Full async API mirrors the sync API.

from agentguard import AsyncTracer, BudgetGuard, patch_openai_async

tracer = AsyncTracer(guards=[BudgetGuard(max_cost_usd=5.00)])
patch_openai_async(tracer)

# All async OpenAI calls are now tracked and budget-enforced

Production: Dashboard + Kill Switch

For teams that need centralized monitoring, alerts, and remote kill switch:

from agentguard import Tracer, HttpSink, BudgetGuard

tracer = Tracer(
    sink=HttpSink(
        url="https://app.agentguard47.com/api/ingest",
        api_key="ag_...",
        batch_size=20,
        flush_interval=10.0,
        compress=True,
    ),
    guards=[BudgetGuard(max_cost_usd=50.00)],
    metadata={"env": "prod"},
    sampling_rate=0.1,  # 10% of traces
)
Free Pro ($39/mo) Team ($79/mo)
SDK + local guards Unlimited Unlimited Unlimited
Dashboard traces - 100K/mo 500K/mo
Budget alerts (email/webhook) - Yes Yes
Remote kill switch - Yes Yes
Team members - 1 10

Start free | View pricing

Architecture

Your Agent Code
    │
    ▼
┌─────────────────────────────────────┐
│           Tracer / AsyncTracer       │  ← trace(), span(), event()
│  ┌───────────┐  ┌────────────────┐  │
│  │  Guards    │  │  CostTracker   │  │  ← runtime intervention
│  └───────────┘  └────────────────┘  │
└──────────┬──────────────────────────┘
           │ emit(event)
    ┌──────┼──────────┬───────────┐
    ▼      ▼          ▼           ▼
 JsonlFile  HttpSink  OtelTrace  Stdout
  Sink      (gzip,    Sink       Sink
            retry)

What's in this repo

Directory Description License
sdk/ Python SDK — guards, tracing, evaluation, integrations MIT
mcp-server/ MCP server — agents query their own traces MIT
site/ Landing page MIT

Dashboard is in a separate private repo (agent47-dashboard).

Security

  • Zero runtime dependencies — one package, nothing to audit, no supply chain risk
  • OpenSSF Scorecard — automated security analysis on every push
  • CodeQL scanning — GitHub's semantic code analysis on every PR
  • Bandit security linting — Python-specific security checks in CI

Contributing

See CONTRIBUTING.md for dev setup, test commands, and PR guidelines.

License

MIT (BMD PAT LLC)

About

Your AI agent just burned $200. AgentGuard stops it at $5. Runtime cost guardrails for AI agents — budget enforcement, loop detection, kill switch. Zero dependencies, MIT licensed.

Topics

Resources

License

Code of conduct

Contributing

Security policy

Stars

Watchers

Forks

Sponsor this project

Packages

No packages published

Contributors 3

  •  
  •  
  •