Self-Improving Agent
一个面向 Automation 场景的 Agent 技能。原始说明:Captures learnings, errors, and corrections to enable continuous improvement. Use when: (1) A command or operation fails unexpectedly, (2) User corrects Clau...
一个面向 Automation 场景的 Agent 技能。原始说明:Discover Snowflake Cortex, Snowpark, notebook, Streamlit, MCP, and AI-observability assets from the operator's environment, emit canonical agent-bom inventor...
name: agent-bom-discover-snowflake
description: >-
Discover Snowflake Cortex, Snowpark, notebook, Streamlit, MCP, and
AI-observability assets from the operator's environment, emit canonical
agent-bom inventory JSON, and scan it without giving agent-bom long-lived
Snowflake credentials. Use when a user asks to inventory Snowflake AI or
Cortex infrastructure as canonical inventory.
version: 0.88.4
license: Apache-2.0
compatibility: >-
Requires Python 3.11+, agent-bom installed with the snowflake extra, and
operator-controlled Snowflake read-only credentials. Prefer SSO, OAuth, or
key-pair auth over passwords.
metadata:
author: msaad00
homepage: https://github.com/msaad00/agent-bom
source: https://github.com/msaad00/agent-bom
pypi: https://pypi.org/project/agent-bom/
openclaw:
requires:
bins:
env: []
credentials: snowflake-read-only
credentialpolicy: "Use the operator's existing Snowflake SSO, OAuth, or key-pair auth context. Prefer SNOWFLAKEPRIVATEKEYPATH or SNOWFLAKEAUTHENTICATOR over SNOWFLAKEPASSWORD. Do not ask users to paste passwords, private keys, or OAuth tokens into chat."
optional_env:
optional_bins:
emoji: "search"
homepage: https://github.com/msaad00/agent-bom
source: https://github.com/msaad00/agent-bom
license: Apache-2.0
os:
credentialhandling: "Credentials stay in the operator environment. The skill invokes Snowflake discovery locally and writes canonical inventory JSON with sourcetype=skillinvokedpull. agent-bom receives sanitized inventory only when the operator explicitly scans or pushes that inventory."
data_flow: "Operator Snowflake account -> read-only Snowflake queries/API calls -> canonical inventory JSON -> optional local agent-bom inventory scan. No agent-bom-hosted service is required. Credential-like values are redacted before persistence/export."
file_reads:
file_writes:
network_endpoints:
purpose: "Snowflake inventory, Cortex, query history, and AI observability discovery"
auth: true
purpose: "Snowflake regional and organization account endpoints selected by the operator"
auth: true
telemetry: false
persistence: false
privilege_escalation: false
always: false
autonomous_invocation: restricted
Use this skill to collect Snowflake AI and workload inventory as schema-valid
agent-bom inventory. Default to discover-only: write JSON to an
operator-selected path and stop.
read-only roles.
SNOWFLAKE_PASSWORD, private key contents, passphrases, or OAuth tokens.
is the evidence.
python examples/operator_pull/snowflake_inventory_adapter.py \
--account "$SNOWFLAKE_ACCOUNT" \
--user "$SNOWFLAKE_USER" \
--authenticator snowflake_jwt \
--source snowflake-skill-invoked \
--discovery-method skill_invoked_pull \
--output snowflake-inventory.json
Scan only when the operator asks for findings:
agent-bom agents --inventory snowflake-inventory.json --format json --output agent-bom-snowflake-findings.json
The emitted inventory carries discovery_provenance.source_type:,
skill_invoked_pullobserved_via: skill_invoked_pull, snowflake_sdk,
sanitized metadata.permissions_used, and redacted credential material. If
schema validation fails, stop and fix the inventory instead of scanning a
best-effort summary.