Skip to main content

AgentMetrics observability plugin for Hermes agents

Project description

agentmetrics-hermes

PyPI License: MIT

AgentMetrics observability plugin for Hermes agents. Auto-discovered via the hermes_agent.plugins entry-point — no code changes needed.


Install

pip install agentmetrics-hermes

Configuration

Create ~/.config/hermes/agentmetrics.yaml:

api_key: am_your_key
server_url: http://localhost:4000
agent_id: my-hermes-agent
environment: production

Or set environment variables (take precedence over the file):

Variable Description
AGENTMETRICS_API_KEY API key
AGENTMETRICS_SERVER_URL Server URL
AGENTMETRICS_AGENT_ID Agent identifier
AGENTMETRICS_ENVIRONMENT Deployment environment

What gets tracked

Metric Description
input_tokens / output_tokens Aggregated LLM token usage
cache_read_tokens / cache_write_tokens Prompt cache statistics
estimated_cost_usd Computed from model pricing
duration_ms Session wall-clock time
tool_calls / tool_errors Per-step tool tracking
llm_calls Number of model API requests
secrets_blocked_count Credentials redacted before transmission
error Exception message on failure (secrets scrubbed)

Durability

Events are written to a local WAL before transmission. If the server is unreachable the plugin retries automatically via an internal circuit breaker. Events that exhaust all retries go to a dead-letter queue and are surfaced as audit_dlq_alert events.


Security

All event payloads are scanned for API keys, JWTs, and passwords before transmission. Set redaction_mode: debug in the config to disable scrubbing during local development (expires automatically after a configurable TTL).


License

MIT

Project details


Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Source Distribution

agentmetrics_hermes-0.2.0.tar.gz (29.8 kB view details)

Uploaded Source

Built Distribution

If you're not sure about the file name format, learn more about wheel file names.

agentmetrics_hermes-0.2.0-py3-none-any.whl (26.8 kB view details)

Uploaded Python 3

File details

Details for the file agentmetrics_hermes-0.2.0.tar.gz.

File metadata

  • Download URL: agentmetrics_hermes-0.2.0.tar.gz
  • Upload date:
  • Size: 29.8 kB
  • Tags: Source
  • Uploaded using Trusted Publishing? No
  • Uploaded via: twine/6.2.0 CPython/3.13.5

File hashes

Hashes for agentmetrics_hermes-0.2.0.tar.gz
Algorithm Hash digest
SHA256 16534771454d114553f4abce35006e95796b137a0a9938a80d310f3938257291
MD5 10c58c0f3216b87b31a5367852febd68
BLAKE2b-256 438416ed908af60f32bd49f2cde04cbae1fe141b385867985570b0d29df269ed

See more details on using hashes here.

File details

Details for the file agentmetrics_hermes-0.2.0-py3-none-any.whl.

File metadata

File hashes

Hashes for agentmetrics_hermes-0.2.0-py3-none-any.whl
Algorithm Hash digest
SHA256 07bb065157ceb7764d5e06d288d362aa1117b47dd645c71ec74f9921fe907ee9
MD5 71abc98f1f2b144362a7fa2d8ca6cc18
BLAKE2b-256 dbc6c21cd3772aaa09d2b1cc1fb8c4defbf64dc152349584134ef6185364a7f2

See more details on using hashes here.

Supported by

AWS Cloud computing and Security Sponsor Datadog Monitoring Depot Continuous Integration Fastly CDN Google Download Analytics Pingdom Monitoring Sentry Error logging StatusPage Status page