Repository navigation
Expand file tree
/
Copy pathprivate.env.example
More file actions
49 lines (38 loc) · 2.51 KB
/
Copy pathprivate.env.example
File metadata and controls
49 lines (38 loc) · 2.51 KB
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
# Private values: what tracked files need but must not name, because this repo is public. Copy
# this file to private.env, which git ignores, and fill in the keys this machine reads. Every key
# below is read only where `is_work` holds: the work laptop, and an Amp orb, whose missing computer
# name also counts as work. The other Macs need no private.env.
#
# Every new shell, and every bash script started outside one, sources private.env with allexport
# on, so each assignment in it is exported. It is shell code: quote a value that holds spaces. The
# values below are made up.
#
# A token, key or password does not go here, because every process the shell starts would get it.
# Store it with `secret add <name>` instead (see docs/standards/secrets.md).
# read by: tools/gcloud/shell.zsh, as the default gcloud project
PRIVATE_GCLOUD_PROJECT=example-project
# read by: tools/gcloud/shell.zsh, as $DATALAKE and the project `bqq` queries
PRIVATE_BQ_PROJECT=example-datalake-1234
# read by: tools/gcloud/shell.zsh, as $AGGREGATED_MAPS
PRIVATE_BQ_TABLE=example_dataset.example_table
# read by: tools/gcloud/install.bash, for the project of the cluster kubectl connects to
PRIVATE_GKE_PROJECT=example-cluster-1234
# read by: tools/claude/cc.bash, which exports it as ANTHROPIC_VERTEX_PROJECT_ID for every work
# session so the switch in settings.json can move them to Vertex AI, and by `cc-vertex` and
# `cc-switch` in tools/claude/shell.zsh
PRIVATE_VERTEX_PROJECT=example-vertex-1234
# read by: tools/uv/install.bash, for the internal uv installer
PRIVATE_UV_INSTALLER_URL=https://internal.example.com/setup-uv.sh
# read by: the use-jira skill and tools/jira/README.md, as the Jira site's hostname
PRIVATE_JIRA_SITE=example.atlassian.net
# read by: the use-jira skill, as the cloudId the Atlassian MCP tools take
PRIVATE_JIRA_CLOUD_ID=00000000-0000-0000-0000-000000000000
# read by: tools/claude/cc.bash, as space-separated FOLDER_PREFIX=RELATED_REPO pairs: a session
# started in a folder whose name starts with FOLDER_PREFIX is told where RELATED_REPO is cloned
PRIVATE_RELATED_REPOS="example-app=example-infra example-infra=example-app"
# read by: features/auth/auth.py, as the Spacelift account `auth` checks and logs in to, the
# subdomain of its <account>.app.spacelift.io URL
PRIVATE_SPACELIFT_ACCOUNT=example
# read by: features/auth/auth.py, as the space-separated Argo CD servers `auth` checks and logs in
# to; after an Argo CD login, the first becomes argocd's default context
PRIVATE_ARGOCD_SERVERS="cd-prod.example.com cd-dev.example.com"