Removing !developer_secrets from all docs #496
Conversation
Removing instructions for !developer_secrets
|
Note Other AI code review bot(s) detectedCodeRabbit has detected other AI code review bot(s) in this pull request and will avoid duplicating their findings in the review comments. This may lead to a less comprehensive review. WalkthroughDocumentation removes explicit Changes
Sequence Diagram(s)sequenceDiagram
autonumber
actor Dev as Developer
participant CLI as mcp-agent CLI
participant Cloud as MCP Agent Cloud
participant SM as Secrets Manager
participant Agent as Deployed Agent
Dev->>CLI: mcp-agent deploy <app>
CLI->>Cloud: Initiate deployment
Cloud-->>CLI: Ask how to handle secrets
CLI->>Dev: Prompt for secrets (interactive)
Dev-->>CLI: Provide secret values / choose management
CLI->>SM: Store secrets / request handles
SM-->>CLI: Return secret handles
CLI->>Cloud: Submit deployment with secret handles
Cloud->>Agent: Provision and start
Agent-->>Dev: Endpoint / ready
note over CLI,Cloud: Errors surfaced if secret validation fails
Estimated code review effort🎯 3 (Moderate) | ⏱️ ~20 minutes Possibly related PRs
Suggested reviewers
Poem
Pre-merge checks and finishing touches✅ Passed checks (3 passed)
✨ Finishing touches🧪 Generate unit tests
Thanks for using CodeRabbit! It's free for OSS, and your support helps us grow. If you like it, consider giving us a shout-out. Comment |
![graphite-app[bot]](https://avatars.githubusercontent.com/in/158384?s=60&v=4){kind=small}
![coderabbitai[bot]](https://avatars.githubusercontent.com/in/347564?s=60&v=4){kind=small}
There was a problem hiding this comment.
Actionable comments posted: 0
🧹 Nitpick comments (6)
examples/usecases/mcp_github_to_slack_agent/README.md (1)
128-140: Fix tool description and grammar.
- “workflows-resume: signal workflow to pause run” should say “resume run”.
- “Tool's that your agent expose” → “Tools that your agent exposes”.
- - workflows-resume: signal workflow to pause run + - workflows-resume: resume a paused run - - Tool's that your agent expose: - - github_to_slack: default of your tool name, input the parameters to trigger a workflow run + - Tools that your agent exposes: + - github_to_slack: default tool name; provide parameters to trigger a workflow runexamples/mcp_agent_server/asyncio/README.md (2)
159-176: Add languages to fenced code blocks (MD040).Mark code fences with languages for login/output/deploy snippets.
-``` +```bash uv run mcp-agent login-
+text
andrew_lm@Mac sdk-cloud % uv run mcp-agent login
INFO: Directing to MCP Agent Cloud API login...
Please enter your API key 🔑:-``` +```bash uv run mcp-agent deploy mcp_agent_server -c /absolute/path/to/your/project--- `178-179`: **Typo: “ou” → “you”.** ```diff -5. In the terminal, ou will then be prompted to specify your OpenAI and/or Anthropic keys: +5. In the terminal, you will then be prompted to specify your OpenAI and/or Anthropic keys:docs/cloud/getting-started.mdx (2)
72-76: Avoid implying a secrets “file” in the deploy stepsWording still points users to a file-based flow. Align this with the new interactive secrets flow.
Apply:
- 2. Process your secrets file + 2. Handle secrets via the interactive deployment flow
320-323: Remove lingering reference to mcp_agent.secrets.yaml and tighten phrasingThis section still mentions a secrets YAML file, which conflicts with the PR goal of moving away from file-based guidance.
Within the changed lines, update bullets:
-1. During `mcp-agent deploy`, you will be prompted through the secrets management flow -2. Secrets are safely secured in MCP Agent Cloud +1. During `mcp-agent deploy`, you’ll be guided through an interactive secrets flow. +2. Secrets are securely stored and injected by MCP Agent Cloud at runtime.Additionally, update the section lead-in (Line 320) to remove the file reference:
### Secrets Management MCP Agent Cloud securely handles secrets during deployment:docs/cloud/overview.mdx (1)
56-57: Good addition; consider cross-linking and consistent phrasingThe new note looks good. For consistency with Getting Started, consider: “During deploy, you’ll be guided through an interactive secrets flow.” Optionally add a link to the “Secrets Management” section in Getting Started.
📜 Review details
Configuration used: CodeRabbit UI
Review profile: CHILL
Plan: Pro
📒 Files selected for processing (12)
docs/cloud/agent-server.mdx(0 hunks)docs/cloud/getting-started.mdx(2 hunks)docs/cloud/overview.mdx(1 hunks)docs/configuration.mdx(0 hunks)examples/basic/mcp_basic_agent/README.md(1 hunks)examples/basic/mcp_model_selector/README.md(1 hunks)examples/mcp_agent_server/asyncio/README.md(1 hunks)examples/usecases/mcp_github_to_slack_agent/README.md(1 hunks)examples/workflows/workflow_evaluator_optimizer/README.md(1 hunks)examples/workflows/workflow_intent_classifier/README.md(1 hunks)examples/workflows/workflow_orchestrator_worker/README.md(1 hunks)src/mcp_agent/cli/README.md(1 hunks)
💤 Files with no reviewable changes (2)
- docs/configuration.mdx
- docs/cloud/agent-server.mdx
🧰 Additional context used
🪛 markdownlint-cli2 (0.17.2)
examples/mcp_agent_server/asyncio/README.md
159-159: Fenced code blocks should have a language specified
(MD040, fenced-code-language)
167-167: Fenced code blocks should have a language specified
(MD040, fenced-code-language)
174-174: Fenced code blocks should have a language specified
(MD040, fenced-code-language)
🔇 Additional comments (6)
examples/workflows/workflow_orchestrator_worker/README.md (1)
79-86: LGTM on removing explicit developer-secrets YAML and adding deployment-time secrets note.The updated flow is clearer and consistent with the PR objective.
examples/basic/mcp_basic_agent/README.md (1)
99-106: LGTM on removing developer-secrets YAML and clarifying deployment-time secrets.The new flow is concise and aligns with the PR objective.
examples/basic/mcp_model_selector/README.md (1)
75-83: LGTM on secrets guidance update and connect section.Consistent with the interactive secrets approach.
examples/workflows/workflow_intent_classifier/README.md (1)
63-71: LGTM on removing developer-secrets YAML and adding deployment-time note.Matches the PR intent without losing necessary context.
src/mcp_agent/cli/README.md (1)
41-41: Remove remaining !developer_secret usage and align README examples with the PR objective.README still shows !developer_secret in the secrets YAML; remove the developer-secret YAML block, clarify that developer secrets are created via the Secrets API or during deployment, and keep !user_secret placeholders.
Location: src/mcp_agent/cli/README.md (around lines 41, 70–77, 79–88).Apply:
-```yaml -# mcp_agent.secrets.yaml (separate secrets file) -api: - key: !developer_secret API_KEY # Developer provides value through API_KEY environment variable - -database: - # User will provide this later - no environment variable specified - password: !user_secret -``` +> Note: Developer secrets are created via the Secrets API or during deployment prompts. Do not store developer secret values in a local YAML file. + +```yaml +# mcp_agent.secrets.yaml (optional: user placeholders only) +database: + password: !user_secret +```Optional paragraph tweak: “When processed during deployment, developer secrets are referenced by UUIDs and user placeholders remain tags.”
examples/workflows/workflow_evaluator_optimizer/README.md (1)
154-167: Standardize tool names with other docs — confirm canonical CLI namesREADME (examples/workflows/workflow_evaluator_optimizer/README.md lines 154–167) uses singular tool names (
workflow-list,workflow-run-list,workflow-run) while other docs use plural. Provide the authoritative tool names from the CLI/help (copy the CLI tool listing or --help output) and align all docs accordingly.
Co-authored-by: graphite-app[bot] <96075541+graphite-app[bot]@users.noreply.github.com>
There was a problem hiding this comment.
Actionable comments posted: 1
Caution
Some comments are outside the diff and can’t be posted inline due to platform limitations.
⚠️ Outside diff range comments (1)
examples/mcp_agent_server/asyncio/README.md (1)
183-207: Cloud output still references “Secrets file” and “Developer secret handle”—contradicts PR goal to remove developer_secrets.Sanitize the deployment transcript to avoid implying secrets.yaml/developer-secret flows in cloud.
Apply this diff to redact the developer-secrets bits while keeping the gist of a successful deploy:
andrew_lm@Mac sdk-cloud % uv run mcp-agent deploy basic_agent_server -c /Users/andrew_lm/Documents/GitHub/mcp-agent/examples/mcp_agent_server/asyncio/ ╭─────────────────────────────────────────────────── MCP Agent Deployment ────────────────────────────────────────────────────╮ │ Configuration: /Users/andrew_lm/Documents/GitHub/mcp-agent/examples/mcp_agent_server/asyncio/mcp_agent.config.yaml │ -│ Secrets file: /Users/andrew_lm/Documents/GitHub/mcp-agent/examples/mcp_agent_server/asyncio/mcp_agent.secrets.yaml │ │ Mode: DEPLOY │ ╰──────────────────────────────────────────────────────── LastMile AI ────────────────────────────────────────────────────────╯ INFO: Using API at https://mcp-agent.com/api INFO: Checking for existing app ID for 'basic_agent_server'... SUCCESS: Found existing app with ID: app_dd3a033d-4f4b-4e33-b82c-aad9ec43c52f for name 'basic_agent_server' -INFO: Processing secrets file... -INFO: Found existing transformed secrets to use where applicable: -/Users/andrew_lm/Documents/GitHub/mcp-agent/examples/mcp_agent_server/asyncio/mcp_agent.deployed.secrets.yaml -INFO: Loaded existing secrets configuration for reuse -INFO: Reusing existing developer secret handle at 'openai.api_key': mcpac_sc_83d412fd-083e-4174-89b4-ecebb1e4cae9 -INFO: Transformed config written to /Users/andrew_lm/Documents/GitHub/mcp-agent/examples/mcp_agent_server/asyncio/mcp_agent.deployed.secrets.yaml - - Secrets Processing Summary -┏━━━━━━━━━━━┳━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━┓ -┃ Type ┃ Path ┃ Handle/Status ┃ Source ┃ -┡━━━━━━━━━━━╇━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━┩ -│ Developer │ openai.api_key │ mcpac_sc...b1e4qwe9 │ ♻️ Reused │ -└───────────┴────────────────┴─────────────────────┴──────────┘ - -Summary: 0 new secrets created, 1 existing secrets reused -SUCCESS: Secrets file processed successfully -INFO: Transformed secrets file written to /Users/andrew_lm/Documents/GitHub/mcp-agent/examples/mcp_agent_server/asyncio/mcp_agent.deployed.secrets.yaml +INFO: Capturing API keys via interactive prompts... +SUCCESS: Secrets configured for deployment ╭───────────────────────────────────────── Deployment Ready ───────────────────────────────────────────────╮ │ Ready to deploy MCP Agent with processed configuration │ ╰──────────────────────────────────────────────────────────────────────────────────────────────────────────╯ WARNING: Found a __main__ entrypoint in main.py. This will be ignored in the deployment. ▰▰▰▰▰▰▱ ✅ Bundled successfully ▹▹▹▹▹ Deploying MCP App bundle...INFO: App ID: app_ddde033d-21as-fe3s-b82c-aaae4243c52f INFO: App URL: https://770xdsp22y321prwv9rasdfasd9l5zj5.deployments.mcp-agent.com INFO: App Status: OFFLINE ▹▹▹▹▹ ✅ MCP App deployed successfully!
♻️ Duplicate comments (1)
examples/mcp_agent_server/asyncio/README.md (1)
178-179: Typo fix confirmed (“you”)—resolved.Matches the prior review note.
🧹 Nitpick comments (6)
examples/mcp_agent_server/asyncio/README.md (6)
159-161: Add a language to the fenced code block (markdownlint MD040).Use bash for CLI commands.
Apply this diff:
-``` +```bash uv run mcp-agent login--- `167-171`: **Label transcript block with a language (markdownlint MD040).** Use console for interactive prompts. Apply this diff: ```diff -``` +```console andrew_lm@Mac sdk-cloud % uv run mcp-agent login INFO: Directing to MCP Agent Cloud API login... Please enter your API key 🔑:--- `174-176`: **Add language to deploy command block (markdownlint MD040).** Use bash. Apply this diff: ```diff -``` +```bash uv run mcp-agent deploy mcp_agent_server -c /absolute/path/to/your/project--- `98-114`: **Clarify that secrets.yaml is optional for local dev; cloud deploy prompts for keys.** Prevents confusion with the removal of developer_secrets. Apply this diff: ```diff -### API Keys +### API keys (local development) @@ -1. Copy the example secrets file: +1. (Optional) Copy the example secrets file for local runs: @@ -2. Edit `mcp_agent.secrets.yaml` to add your API keys: +2. (Optional) Edit `mcp_agent.secrets.yaml` to add your API keys: @@anthropic:
api_key: "your-anthropic-api-key"
openai:
api_key: "your-openai-api-key"+ +Note: For cloud deployments, you will be prompted to enter keys interactively during `mcp-agent deploy`; a local secrets file is not required.
25-37: “Synchronous tool” example is declared async; align code and imports.Make it truly sync and import Context.
Apply this diff:
-from mcp_agent.app import MCPApp +from mcp_agent.app import MCPApp, Context @@ -# Synchronous tool – returns the final result to the caller -@app.tool -async def grade_story(story: str, app_ctx: Optional[Context] = None) -> str: +# Synchronous tool – returns the final result to the caller +@app.tool +def grade_story(story: str, app_ctx: Optional[Context] = None) -> str: @@ - return "Report..." + return "Report..."If
Contextis not a public type here, switch the annotation toOptional[object]instead.
356-357: Mark secrets file in Code Structure as optional.Keeps docs consistent with interactive secrets in cloud.
Apply this diff:
-- `mcp_agent.secrets.yaml` - Contains API keys (not included in repository) +- `mcp_agent.secrets.yaml` - Optional (local dev): contains API keys (not included in repository)
📜 Review details
Configuration used: CodeRabbit UI
Review profile: CHILL
Plan: Pro
📒 Files selected for processing (1)
examples/mcp_agent_server/asyncio/README.md(1 hunks)
🧰 Additional context used
🪛 markdownlint-cli2 (0.17.2)
examples/mcp_agent_server/asyncio/README.md
159-159: Fenced code blocks should have a language specified
(MD040, fenced-code-language)
167-167: Fenced code blocks should have a language specified
(MD040, fenced-code-language)
174-174: Fenced code blocks should have a language specified
(MD040, fenced-code-language)
🔇 Additional comments (1)
examples/mcp_agent_server/asyncio/README.md (1)
158-166: Login flow reorder reads better—LGTM.The authentication and API-key setup steps are clearer and match the interactive flow.
| ### [Beta] Deploying to mcp-agent cloud | ||
|
|
||
| You can deploy your MCP-Agent app as a hosted mcp-agent app in the Cloud. | ||
|
|
||
| 1. Update the *mcp_agent.secrets.yaml* file specifying deployed secrets: | ||
|
|
||
| ``` | ||
| openai: | ||
| api_key: !developer_secret | ||
| anthropic: | ||
| api_key: !developer_secret | ||
| ``` | ||
|
|
||
| 2. In your terminal, authenticate into mcp-agent cloud by running: | ||
| 1. In your terminal, authenticate into mcp-agent cloud by running: | ||
| ``` | ||
| uv run mcp-agent login | ||
| ``` | ||
|
|
||
| 3. You will be redirected to the login page, create an mcp-agent cloud account through Google or Github | ||
| 2. You will be redirected to the login page, create an mcp-agent cloud account through Google or Github | ||
|
|
||
| 4. Set up your mcp-agent cloud API Key and copy & paste it into your terminal | ||
| 3. Set up your mcp-agent cloud API Key and copy & paste it into your terminal | ||
|
|
||
| ``` | ||
| andrew_lm@Mac sdk-cloud % uv run mcp-agent login | ||
| INFO: Directing to MCP Agent Cloud API login... | ||
| Please enter your API key 🔑: | ||
| ``` | ||
|
|
||
| 5. In your terminal, deploy the MCP app: | ||
| 4. In your terminal, deploy the MCP app: | ||
| ``` | ||
| uv run mcp-agent deploy mcp_agent_server -c /absolute/path/to/your/project | ||
| ``` | ||
|
|
||
| 6. In the terminal, ou will then be prompted to specify your OpenAI and/or Anthropic keys: | ||
| 5. In the terminal, you will then be prompted to specify your OpenAI and/or Anthropic keys: | ||
|
|
||
| Once the deployment is successful, you should see the following: | ||
| ``` |
There was a problem hiding this comment.
🛠️ Refactor suggestion
🧩 Analysis chain
Add a quick self-check to keep MD040 clean going forward.
Run markdownlint locally to verify no remaining MD040 in this file.
🏁 Script executed:
#!/bin/bash
# Verify fenced code blocks have languages
npx -y markdownlint-cli2 'examples/mcp_agent_server/asyncio/README.md' || true
rg -n '```$' examples/mcp_agent_server/asyncio/README.mdLength of output: 6704
Add markdownlint self-check and fix MD040 occurrences
markdownlint-cli2 found 50 errors in examples/mcp_agent_server/asyncio/README.md, including multiple MD040 (fenced code blocks missing language) — e.g. lines 102, 108, 121, 139, 159, 174, 181, 256, 293, 333. Action items:
- Add language tags to every ``` fence and fix the other lint issues (line-length, list numbering, blank lines).
- Add a CI or pre-commit check running npx -y markdownlint-cli2 and require no MD040 before merging; re-run the linter until clean.
🧰 Tools
🪛 markdownlint-cli2 (0.17.2)
159-159: Fenced code blocks should have a language specified
(MD040, fenced-code-language)
167-167: Fenced code blocks should have a language specified
(MD040, fenced-code-language)
174-174: Fenced code blocks should have a language specified
(MD040, fenced-code-language)
181-181: Fenced code blocks should have a language specified
(MD040, fenced-code-language)
🤖 Prompt for AI Agents
In examples/mcp_agent_server/asyncio/README.md around lines 155 to 181, several
fenced code blocks lack language tags (MD040) and other markdownlint issues
exist; add appropriate language identifiers (bash, sh, or console) to every ```
fence (including at lines 102, 108, 121, 139, 159, 174, 181, etc.), fix other
lint problems such as line-length, list numbering, and required blank lines per
markdownlint rules, and add a CI/pre-commit step that runs npx -y
markdownlint-cli2 (failing on any MD040) so the linter must pass before merging.
| Two types of secrets are supported: | ||
|
|
||
| 1. **Developer Secrets** (`!developer_secret`): | ||
| 1. **Developer Secrets**: |
There was a problem hiding this comment.
I didn't see your PR before I did the same. Your's is good, but we should also make some other changes in this README (e.g. see main...fix/update-docs-for-secrets-tags) and should delete src/mcp_agent/cli/SECRETS.md entirely to avoid confusion
3 participants
Removing instructions for !developer_secrets
Summary by CodeRabbit
Documentation
Style