`cf workers secrets update` replaces the entire secret set instead of patching a single secret
Maintainers usually reply within 1 day
Nobody has claimed this yet.
Assessment
- Difficulty
- 3/5
- Estimated time
- 1-2 days
- Newbie friendliness
- 68/100
- Issue type
- Bug
- Clarity
- Clearly specified
- Activity status
- Active
- Tech stack
- typescript
Research direction
Start at the cf workers secrets update command entry point and compare its dry-run request with cf workers secrets bulk, including the /secrets and /secrets-bulk endpoints. Reproduce the two-secret sequence from the issue and inspect the existing command tests or request handling. Done means updating one secret leaves other secrets unchanged, with coverage for the reported sequence or clearly documented replacement behavior.
Written by the indexing model from the issue text.
Description
cf version: 1.0.0-beta.5, macOS
What happens
cf workers secrets update <NAME> --worker <worker> --type secret_text --text <value> sets the named secret, but silently deletes every other secret already on the Worker — it's not a patch, it's a full replace.
Hit this for real while migrating a production Worker (8 secrets already set via wrangler secret put, working fine). Ran cf workers secrets update RUN_TOKEN --worker web-qa-jev-agent --type secret_text --text "<new-token>" to rotate one token. cf workers secrets list --worker web-qa-jev-agent immediately after came back [{"name": "RUN_TOKEN", ...}] — the other 7 (GitHub App auth, webhook secret, target URL) were gone. Broke GitHub App auth and a webhook trigger until manually restored via wrangler secret put.
Repro
cf init repro-cf --package-manager npm
cd repro-cf
cf deploy # after declaring at least one bindings.secret() per #70/#72's workaround, or use --secrets-file
cf workers secrets update FIRST --worker repro-cf --type secret_text --text a
cf workers secrets list --worker repro-cf # [FIRST]
cf workers secrets update SECOND --worker repro-cf --type secret_text --text b
cf workers secrets list --worker repro-cf # [SECOND] -- FIRST is gone
What I found
cf workers secrets update --dry-run shows exactly why:
{
"command": "cf workers secrets update",
"method": "PUT",
"url": "https://api.cloudflare.com/client/v4/accounts/{account}/workers/scripts/{script}/secrets",
"body": { "name": "SECOND", "type": "secret_text", "text": "b" }
}
It's a PUT to the plain /secrets collection endpoint with only the one secret in the body — a full replace, not a merge. Compare to cf workers secrets bulk --dry-run with the same secret:
{
"command": "cf workers secrets bulk",
"method": "PATCH",
"url": "https://api.cloudflare.com/client/v4/accounts/{account}/workers/scripts/{script}/secrets-bulk",
"body": { "SECOND": { "type": "secret_text", "text": "b" } }
}
bulk hits a different endpoint (/secrets-bulk) with PATCH/RFC 7396 merge semantics — its own --help text even says "Secrets not included in the request are left unchanged" and recommends it over update for multi-secret changes. But nothing in update's own --help or command name suggests that a single-secret call is just as destructive to every other existing secret. This looks like the same root-cause family as #70/#72/#54 (PUT-only API, no merge-on-single-field-update), specifically for the one command whose entire job description ("Add a secret to a Worker script") implies it should be additive.
Expected
cf workers secrets update either performs a genuine patch (e.g. by using the bulk/secrets-bulk endpoint under the hood when only one secret is given), or its --help/docs state unambiguously that it replaces the whole secret set — right now the command name and description ("Add a secret...") actively suggest the opposite of what it does.
Workaround
Use cf workers secrets bulk --body '{"NAME": {"type": "secret_text", "text": "value"}}' instead of update, always, even for a single secret on a Worker that has others set.
- Dominant language
- TypeScript
- Stars
- 853
- Forks
- 42
- Avg merge
- 17h 57m
- Merged PRs (30d)
- 146
Getting set up
- No Dockerfile or Docker Compose file
- No pull request template
- Read the contributing guide
First steps
- Read the whole issue, then the project's contributing guide.
- Comment on the issue to say you are picking it up — it saves two people doing the same work.
- Fork the repository and make your change on a branch.
- Open a pull request that references the issue number.
More from cloudflare/cf
-
Difficulty 2/5 1-3 hours Newbie friendliness 78/100
cloudflare/cf#344 ·
Maintainers usually reply within 1 day
-
upstream
Difficulty 2/5 1-3 hours Newbie friendliness 72/100
cloudflare/cf#262 ·
Maintainers usually reply within 1 day
-
`cf dev` rejects `--port` and `--host` for Vite projects, including the `cf init` templatePossibly taken @jamesopstad claimed this 6 days ago. Open
Difficulty 2/5 1-3 hours Newbie friendliness 72/100
cloudflare/cf#223 · 2 assignees ·
Maintainers usually reply within 1 day
-
upstream
Difficulty 2/5 1-3 hours Newbie friendliness 75/100
cloudflare/cf#179 ·
Maintainers usually reply within 1 day
-
Difficulty 4/5 3-5 days Newbie friendliness 58/100
cloudflare/cf#347 ·
Maintainers usually reply within 1 day
Similar issues
-
Difficulty 2/5 1-3 hours Newbie friendliness 78/100
jessepollak/home#2223 ·
Maintainers usually reply within 1 day
-
grep fixture corpus test follows the corpus's loop symlink under the receipt runner (fails on main)Open
Difficulty 2/5 1-3 hours Newbie friendliness 68/100
code-yeongyu/senpi#3756 ·
Maintainers usually reply within 1 day
-
org:external package:deepagents priority:triage topic:memory topic:middleware type:bug
Difficulty 2/5 1-3 hours Newbie friendliness 65/100
langchain-ai/deepagents#7059 ·
Maintainers usually reply within 1 day
-
Difficulty 1/5 Under an hour Newbie friendliness 92/100
osmlab/name-suggestion-index#12992 ·
Maintainers usually reply within 1 day
-
bug
Difficulty 2/5 1-3 hours Newbie friendliness 66/100
hemilabs/ui-monorepo#2506 ·
Maintainers usually reply within 1 day