agentic-ledger 0.1.1

Runtime observability for AI agents — see exactly what your agent did, why it did it, and what it cost.

AgentLedger

Runtime observability for AI agents — see exactly what your agent did, why it did it, and what it cost.

Works with any agent framework, any LLM provider, any model gateway. Zero code changes required. Point your agent at the proxy and everything is captured automatically.

---

How it works

AgentLedger runs as a transparent proxy between your agent and the LLM provider. It intercepts every request and response, assigns it an action_id, stores it, and returns the upstream response unmodified. Your agent never knows the proxy is there.

Your Agent  →  AgentLedger Proxy  →  OpenAI / Anthropic / LiteLLM / any LLM
                      ↓
               SQLite or Postgres
                      ↓
               Live Dashboard + API

---

Quick Start

Step 1 — Start the proxy

With Docker (recommended, no Python required):

docker run -p 8000:8000 \
  -e AGENTLEDGER_UPSTREAM_URL=https://api.openai.com \
  -v $(pwd)/data:/data \
  ghcr.io/shekharbhardwaj/agentledger:latest

Or with docker compose (SQLite by default, Postgres available — see docker-compose.yml):

AGENTLEDGER_UPSTREAM_URL=https://api.openai.com docker compose up

With uv:

uv add agentic-ledger
AGENTLEDGER_UPSTREAM_URL=https://api.openai.com uv run python -m agentledger.proxy

With pip:

python -m venv venv && source venv/bin/activate
pip install agentic-ledger
AGENTLEDGER_UPSTREAM_URL=https://api.openai.com ./venv/bin/python -m agentledger.proxy

Postgres? Install the extra and set AGENTLEDGER_DSN:

pip install "agentic-ledger[postgres]"
AGENTLEDGER_DSN=postgresql://user:password@localhost/agentledger

OpenTelemetry? Install the extra and set AGENTLEDGER_OTEL_ENDPOINT:

pip install "agentic-ledger[otel]"
AGENTLEDGER_OTEL_ENDPOINT=http://localhost:4318

Proxy starts on http://localhost:8000. Traces are saved to agentledger.db in the current folder (or /data/agentledger.db in Docker).

---

Step 2 — Point your agent at the proxy

Two changes: set base_url to the proxy and add a session ID header to group calls into a run. Everything else — your API key, model, messages — stays exactly the same.

OpenAI:

from openai import OpenAI

client = OpenAI(
    base_url="http://localhost:8000/v1",  # ← proxy
    api_key="your-openai-key",
    default_headers={"x-agentledger-session-id": "run-1"},
)

response = client.chat.completions.create(
    model="gpt-4o",
    messages=[{"role": "user", "content": "Research the top 3 AI trends in 2026"}],
)

Anthropic:

import anthropic

client = anthropic.Anthropic(
    base_url="http://localhost:8000",  # ← proxy
    api_key="your-anthropic-key",
    default_headers={"x-agentledger-session-id": "run-1"},
)

LiteLLM / OpenRouter / any gateway:

# Point AgentLedger at your gateway
AGENTLEDGER_UPSTREAM_URL=http://localhost:4000 uv run python -m agentledger.proxy

# Then point your agent at AgentLedger
client = OpenAI(base_url="http://localhost:8000/v1", ...)

---

Step 3 — Open the dashboard

http://localhost:8000

The dashboard updates live via WebSocket as calls come in. No refresh needed.

• Calls tab — every LLM call with full prompt, system prompt, tool calls, tool results, output, tokens, cost, latency, and errors
• Flow tab — visual DAG of your multi-agent system. Each agent is a node with aggregated cost, latency, and call count. Edges represent handoffs. Click a node to highlight its calls.
• Search — full-text search across all sessions by prompt, output, agent name, or user ID

---

What gets captured

Every LLM call is stored with:

Field	What it contains
action_id	UUID assigned at interception time
session_id	Run grouping (from header)
timestamp	When the call was made
model_id	Model used
provider	openai or anthropic
messages	Full message history sent to the model
system_prompt	Extracted system prompt
tools	Tool definitions available to the model
tool_calls	Tools the model decided to call
tool_results	What the tools returned (from next call's messages)
content	Model's text output
stop_reason	Why the model stopped
tokens_in / tokens_out	Token usage
cost_usd	Estimated cost based on model pricing
latency_ms	End-to-end response time
status_code	HTTP status from upstream — errors are captured too
error_detail	Upstream error message for non-200 responses
agent_name	From x-agentledger-agent-name header
user_id	From x-agentledger-user-id header
app_id	From x-agentledger-app-id header
environment	From x-agentledger-environment header
parent_action_id	Parent call in a nested agent graph
handoff_from / handoff_to	Agent handoff tracking for the Flow DAG

---

API reference

Method	Endpoint	Description
GET	/	Live dashboard
WS	/ws	WebSocket stream — powers live dashboard updates
GET	/api/sessions	List recent sessions with aggregated stats
GET	/api/search?q=...	Full-text search across all captured calls
GET	/session/{session_id}	All calls in a session, ordered by time
GET	/explain/{action_id}	Single call by action ID
GET	/export/{session_id}	JSON compliance export with SHA-256 integrity hash
GET	/export/{session_id}/report	Printable HTML audit report
POST	/mcp	MCP tool server — exposes captured data to other agents

Examples:

# All calls in a session
curl http://localhost:8000/session/run-1

# Search across all sessions
curl "http://localhost:8000/api/search?q=failed+to+connect"

# Download JSON audit trail (includes SHA-256 hash for tamper detection)
curl http://localhost:8000/export/run-1 -o audit-run-1.json

# Printable HTML report — open in browser, print to PDF
open http://localhost:8000/export/run-1/report

---

Configuration

Environment variables

Core:

Variable	Required	Default	Description
AGENTLEDGER_UPSTREAM_URL	Yes	https://api.openai.com	LLM endpoint to forward requests to. Accepts OpenAI, Anthropic, LiteLLM, OpenRouter, or any OpenAI-compatible URL.
AGENTLEDGER_DSN	No	sqlite:///agentledger.db	Database. SQLite for local dev, Postgres URL for production.
AGENTLEDGER_HOST	No	0.0.0.0	Host to bind to. Use 127.0.0.1 to restrict to localhost only.
AGENTLEDGER_PORT	No	8000	Port to run on.
AGENTLEDGER_API_KEY	No	(none)	Protects the dashboard and all read endpoints. Skip for local dev. Set when the proxy is on a server — you choose the value.

Cost budgets — block calls that exceed a spend limit (returns HTTP 429):

Variable	Default	Description
AGENTLEDGER_BUDGET_SESSION	(none)	Max USD per session_id across its lifetime.
AGENTLEDGER_BUDGET_AGENT	(none)	Max USD per agent_name per calendar day (UTC).
AGENTLEDGER_BUDGET_DAILY	(none)	Max USD total across all calls per calendar day (UTC).

Rate limits — block calls that exceed request frequency (returns HTTP 429, sliding 60-second window):

Variable	Default	Description
AGENTLEDGER_RATE_LIMIT_RPM	(none)	Max requests per minute globally.
AGENTLEDGER_RATE_LIMIT_SESSION_RPM	(none)	Max requests per minute per session_id.
AGENTLEDGER_RATE_LIMIT_AGENT_RPM	(none)	Max requests per minute per agent_name.
AGENTLEDGER_RATE_LIMIT_USER_RPM	(none)	Max requests per minute per user_id.

Alerts — POST to your webhook when a threshold is breached (does not block calls — see Alerts):

Variable	Default	Description
AGENTLEDGER_ALERT_WEBHOOK_URL	(none)	URL to POST alert payloads to. Required for any alerts to fire.
AGENTLEDGER_ALERT_COST_PER_CALL	(none)	Alert when a single call costs more than $X.
AGENTLEDGER_ALERT_LATENCY_MS	(none)	Alert when a single call takes longer than Xms.
AGENTLEDGER_ALERT_ERROR_RATE	(none)	Alert when session error rate exceeds X (e.g. 0.5 = 50%).
AGENTLEDGER_ALERT_DAILY_SPEND	(none)	Alert when daily spend crosses $X. Unlike budgets, this does not block calls.

---

Common startup examples

# Local dev — OpenAI (default)
AGENTLEDGER_UPSTREAM_URL=https://api.openai.com uv run python -m agentledger.proxy

# Local dev — Anthropic
AGENTLEDGER_UPSTREAM_URL=https://api.anthropic.com uv run python -m agentledger.proxy

# Local dev — LiteLLM gateway (any model)
AGENTLEDGER_UPSTREAM_URL=http://localhost:4000 uv run python -m agentledger.proxy

# Production — Postgres + auth + budgets + rate limits + alerts
AGENTLEDGER_UPSTREAM_URL=https://api.openai.com \
AGENTLEDGER_DSN=postgresql://user:password@localhost/agentledger \
AGENTLEDGER_API_KEY=my-secret \
AGENTLEDGER_BUDGET_DAILY=20.00 \
AGENTLEDGER_BUDGET_SESSION=2.00 \
AGENTLEDGER_RATE_LIMIT_SESSION_RPM=20 \
AGENTLEDGER_RATE_LIMIT_USER_RPM=60 \
AGENTLEDGER_ALERT_WEBHOOK_URL=https://hooks.slack.com/services/xxx/yyy/zzz \
AGENTLEDGER_ALERT_COST_PER_CALL=0.50 \
AGENTLEDGER_ALERT_DAILY_SPEND=15.00 \
uv run python -m agentledger.proxy

When AGENTLEDGER_API_KEY is set, pass it to access protected endpoints:

# Header
curl -H "x-agentledger-api-key: my-secret" http://localhost:8000/session/run-1

# Query param (browser)
http://localhost:8000?api_key=my-secret

---

Request headers

Pass these from your agent on each LLM call. All optional. They enrich captured data, power the Flow tab, and enable per-dimension budgets and rate limits.

Header	Default	Description
x-agentledger-session-id	(none)	Groups all calls in a run. Use a consistent ID per agent execution (e.g. a UUID or "run-1"). Without this, calls are stored but not grouped in the dashboard.
x-agentledger-user-id	(none)	End user who triggered this run. Enables per-user rate limiting and auditing.
x-agentledger-agent-name	(none)	Name of the agent making this call (e.g. "orchestrator", "researcher"). Powers the Flow tab DAG and agent-level budgets and rate limits.
x-agentledger-app-id	(none)	Application name or ID. Useful when multiple apps share one proxy.
x-agentledger-parent-action-id	(none)	The action_id of the call that spawned this one. Builds the nested agent call graph.
x-agentledger-environment	development	production, staging, or development. Shown in the dashboard.
x-agentledger-handoff-from	(none)	Agent handing off control (e.g. "orchestrator"). Renders as a directed edge in the Flow DAG.
x-agentledger-handoff-to	(none)	Agent receiving control (e.g. "researcher"). Renders as a directed edge in the Flow DAG.

Single agent — fully annotated:

from openai import OpenAI

client = OpenAI(
    base_url="http://localhost:8000/v1",
    api_key="your-openai-key",
    default_headers={
        "x-agentledger-session-id":  "run-abc123",
        "x-agentledger-user-id":     "user-42",
        "x-agentledger-agent-name":  "researcher",
        "x-agentledger-app-id":      "my-app",
        "x-agentledger-environment": "production",
    },
)

Multi-agent system — tracking handoffs:

from openai import OpenAI

# Orchestrator
orchestrator_client = OpenAI(
    base_url="http://localhost:8000/v1",
    api_key="your-openai-key",
    default_headers={
        "x-agentledger-session-id":  "run-abc123",
        "x-agentledger-agent-name":  "orchestrator",
    },
)

# Researcher (receives handoff from orchestrator)
researcher_client = OpenAI(
    base_url="http://localhost:8000/v1",
    api_key="your-openai-key",
    default_headers={
        "x-agentledger-session-id":   "run-abc123",
        "x-agentledger-agent-name":   "researcher",
        "x-agentledger-handoff-from": "orchestrator",
        "x-agentledger-handoff-to":   "researcher",
    },
)

The Flow tab renders orchestrator → researcher as a DAG with cost and latency on each node.

---

Alerts

AgentLedger fires a POST to your webhook URL when a threshold is breached. You connect it to whatever you already use — Slack, PagerDuty, Discord, email, or a custom endpoint. AgentLedger sends the payload; the integration is on your side.

Payload format:

{
  "type":       "high_cost",
  "message":    "Single call cost $0.1842 exceeded threshold $0.10",
  "value":      0.1842,
  "threshold":  0.10,
  "action_id":  "a1b2c3d4-...",
  "session_id": "run-1",
  "agent_name": "researcher",
  "timestamp":  "2026-04-03T12:00:00+00:00"
}

Alert types:

Type	Triggered when
high_cost	A single call exceeds AGENTLEDGER_ALERT_COST_PER_CALL
high_latency	A single call takes longer than AGENTLEDGER_ALERT_LATENCY_MS
high_error_rate	Session error rate exceeds AGENTLEDGER_ALERT_ERROR_RATE
daily_spend	Daily total spend crosses AGENTLEDGER_ALERT_DAILY_SPEND

Budgets vs alerts:

• Budgets (AGENTLEDGER_BUDGET_*) — block the call before it reaches the LLM. Agent gets HTTP 429.
• Alerts (AGENTLEDGER_ALERT_*) — the call goes through, you get notified after.

Slack — create an Incoming Webhook and point AGENTLEDGER_ALERT_WEBHOOK_URL at it.

PagerDuty — use the Events API v2 URL or a thin adapter that maps type → PagerDuty severity.

Discord — use a Discord channel webhook URL directly.

Custom — any HTTP endpoint that accepts a JSON POST.

---

OpenTelemetry export

AgentLedger can emit every intercepted LLM call as an OTel span to any OTLP-compatible collector: Grafana Tempo, Jaeger, Honeycomb, Datadog, Dynatrace, or any vendor that supports OTLP/HTTP.

Install the extra:

pip install "agentic-ledger[otel]"
# or
uv add "agentic-ledger[otel]"

Configure:

Variable	Default	Description
AGENTLEDGER_OTEL_ENDPOINT	(none)	OTLP/HTTP base URL, e.g. http://localhost:4318. OTel export is disabled when not set.
AGENTLEDGER_OTEL_SERVICE_NAME	agentledger	Value of service.name in the emitted resource.
AGENTLEDGER_OTEL_HEADERS	(none)	Comma-separated key=value pairs for auth headers, e.g. x-honeycomb-team=abc123,x-honeycomb-dataset=llm.

Example — Grafana Tempo:

AGENTLEDGER_UPSTREAM_URL=https://api.openai.com \
AGENTLEDGER_OTEL_ENDPOINT=http://localhost:4318 \
AGENTLEDGER_OTEL_SERVICE_NAME=my-agent \
uv run python -m agentledger.proxy

Example — Honeycomb:

AGENTLEDGER_OTEL_ENDPOINT=https://api.honeycomb.io \
AGENTLEDGER_OTEL_HEADERS=x-honeycomb-team=YOUR_API_KEY,x-honeycomb-dataset=llm-traces \
uv run python -m agentledger.proxy

Span attributes emitted (GenAI semantic conventions):

Attribute	Source
gen_ai.system	Provider (openai / anthropic)
gen_ai.operation.name	Always chat
gen_ai.request.model	Model ID
gen_ai.request.temperature	If set
gen_ai.request.max_tokens	If set
gen_ai.usage.input_tokens	Tokens in
gen_ai.usage.output_tokens	Tokens out
gen_ai.response.finish_reasons	Stop reason
agentledger.action_id	Unique call ID
agentledger.session_id	Run grouping
agentledger.agent_name	From header
agentledger.user_id	From header
agentledger.cost_usd	Estimated cost
agentledger.latency_ms	End-to-end latency
agentledger.environment	From header
agentledger.handoff_from / agentledger.handoff_to	Agent handoffs
http.status_code	HTTP status from upstream

Spans are grouped into traces by session_id — all calls in a session appear as one trace in your backend. Parent-child relationships follow x-agentledger-parent-action-id. Error spans (status_code != 200) are marked with StatusCode.ERROR.

---

Compliance export

Every session can be exported as a signed audit trail — useful for regulated industries, internal audits, or passing traces to external tools.

# Machine-readable JSON with SHA-256 integrity hash
curl http://localhost:8000/export/run-1 -o audit-run-1.json

# Printable HTML — open in browser and print to PDF
open http://localhost:8000/export/run-1/report

The JSON export includes a sha256 hash of the calls array. Recipients can verify the export has not been modified after generation.

---

Releasing

Tagging a version triggers the full release pipeline automatically:

git tag v0.2.0
git push origin v0.2.0

This runs three jobs:

1. Docker — builds and pushes ghcr.io/shekharbhardwaj/agentledger:0.2.0 and :latest to GHCR
2. PyPI — builds and publishes agentic-ledger==0.2.0 to PyPI using trusted publishing (no API token needed)
3. GitHub Release — creates a release with auto-generated changelog from commit messages

First-time PyPI setup (one time only):

1. Go to pypi.org/manage/account/publishing
2. Add a new pending publisher:

   PyPI project name:  agentic-ledger
   Owner:              ShekharBhardwaj
   Repository:         AgentLedger
   Workflow name:      release.yml
   Environment name:   pypi
   

3. Create a pypi environment in GitHub: repo → Settings → Environments → New environment → name it pypi
4. That's it — no secrets needed

---

License

MIT