refactor: migrate 7 MCPs to curl-based subagents#172
Conversation
Replace hetzner, hostinger, serper, dataforseo, ahrefs, gsc, and
outscraper MCP servers with curl-based subagent patterns. This
eliminates always-on MCP context cost while preserving full API
functionality on-demand.
- Rewrite hetzner.md with generic HCLOUD_TOKEN_{PROJECT} curl pattern
- Rewrite serper.md with direct curl API calls
- Add dataforseo.md direct API section
- Create new ahrefs.md curl-based subagent
- Update hostinger, gsc, outscraper frontmatter for curl usage
- Update seo.md agent with ahrefs subagent and MCP migration notes
- Update mcp-servers-config.json.txt template (disabled migrated MCPs)
|
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. WalkthroughThis PR converts multiple MCP-based integrations to direct REST API usage across SEO, hosting, and data-extraction docs, adds Ahrefs REST API docs, and reorganizes MCP server configuration entries to favor remote/tool-specific MCPs and token-based auth. Changes
Sequence Diagram(s)(Skipped — changes are documentation/config rewrites and do not introduce a new multi-component runtime control flow that requires visualization.) Estimated code review effort🎯 3 (Moderate) | ⏱️ ~22 minutes Possibly related PRs
Poem
🚥 Pre-merge checks | ✅ 3✅ Passed checks (3 passed)
✏️ Tip: You can configure your own custom pre-merge checks in the settings. 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 |
Summary of ChangesHello @marcusquinn, I'm Gemini Code Assist1! I'm currently reviewing this pull request and will post my feedback shortly. In the meantime, here's a summary to help you and other reviewers quickly get up to speed! This pull request refactors the integration of several external services by transitioning from persistent MCP servers to on-demand, curl-based subagents. The primary goal is to optimize resource usage and reduce context token consumption by ensuring that API integrations are only active when explicitly called, thereby improving overall efficiency and cost-effectiveness of the system. Highlights
Using Gemini Code AssistThe full guide for Gemini Code Assist can be found on our documentation page, here are some quick tips. Invoking Gemini You can request assistance from Gemini at any point by creating a comment using either
Customization To customize Gemini Code Assist for GitHub experience, repository maintainers can create a configuration file and/or provide a custom code review style guide (such as PEP-8 for Python) by creating and adding files to a Limitations & Feedback Gemini Code Assist may make mistakes. Please leave feedback on any instances where its feedback is incorrect or counter productive. You can react with 👍 and 👎 on @gemini-code-assist comments. If you're interested in giving your feedback about your experience with Gemini Code Assist for Github and other Google products, sign up here. You can also get AI-powered code generation, chat, as well as code reviews directly in the IDE at no cost with the Gemini Code Assist IDE Extension. Footnotes
|
🔍 Code Quality Report�[0;35m[MONITOR]�[0m Code Review Monitoring Report �[0;34m[INFO]�[0m Latest Quality Status: �[0;34m[INFO]�[0m Recent monitoring activity: 📈 Current Quality Metrics
Generated on: Sat Jan 24 14:45:42 UTC 2026 Generated by AI DevOps Framework Code Review Monitoring |
![gemini-code-assist[bot]](https://avatars.githubusercontent.com/in/956858?s=60&v=4){kind=small}
There was a problem hiding this comment.
Code Review
This pull request is a significant refactoring that migrates seven MCP servers to a more efficient curl-based subagent pattern. This change effectively reduces context token usage by eliminating always-on servers, which is a great improvement. The documentation for the new curl-based interactions is clear and comprehensive, with good examples.
My review focuses on improving the robustness of the new curl commands and ensuring consistency between the implementation and the stated design decisions. I've identified a few areas for improvement:
- Consistently using the
Accept: application/jsonheader for API calls. - A potential security improvement for installing packages from Git.
- A discrepancy between the PR description and the changes in the MCP configuration file.
Overall, this is a solid contribution that improves the architecture. The feedback provided should help refine it further.
| - **Database management**: Automated database operations | ||
|
|
||
| --- | ||
| Install: `uv tool install 'mcp-hetzner @ git+https://github.com/dkruyt/mcp-hetzner.git'` |
There was a problem hiding this comment.
Installing packages directly from a Git repository can pose a security risk, as it typically pulls from the latest commit on the default branch, which may not be stable or secure. It is safer to install from a trusted package registry or, if installing from Git, to pin to a specific commit hash or tag for reproducibility and to mitigate the risk of running untrusted code.
.agent/seo/ahrefs.md
Outdated
| curl -s "https://api.ahrefs.com/v3/site-explorer/backlinks?target=example.com&mode=domain&limit=50" \ | ||
| -H "Authorization: Bearer $AHREFS_API_KEY" |
There was a problem hiding this comment.
For consistency and to ensure the server returns JSON, it's a good practice to include the Accept: application/json header in all API calls that are expected to return JSON. This header is present in the 'Site Explorer - Overview' example but is missing here. This advice applies to other new curl examples in this PR as well.
For example:
curl -s "https://api.ahrefs.com/v3/site-explorer/backlinks?target=example.com&mode=domain&limit=50" \
-H "Authorization: Bearer $AHREFS_API_KEY" \
-H "Accept: application/json"| # Edit with your actual API tokens and server details | ||
| # Load token for a specific project | ||
| source ~/.config/aidevops/mcp-env.sh | ||
| export HCLOUD_TOKEN="$HCLOUD_TOKEN_MYPROJECT" |
There was a problem hiding this comment.
The placeholder MYPROJECT is a bit ambiguous. For better clarity and consistency with the examples in the 'Multi-Project Setup' section (lines 158-159), consider using one of those, like PROJECTA.
| export HCLOUD_TOKEN="$HCLOUD_TOKEN_MYPROJECT" | |
| export HCLOUD_TOKEN="$HCLOUD_TOKEN_PROJECTA" |
🤖 Augment PR SummarySummary: Migrates several “always-on” MCP integrations to curl-driven subagent docs so API capabilities are loaded on-demand and reduce per-session context cost. Changes:
Technical Notes: Tokens/credentials are expected to live in 🤖 Was this summary useful? React with 👍 or 👎 |
![augmentcode[bot]](https://avatars.githubusercontent.com/in/1027498?s=60&v=4){kind=small}
| - gsc-sitemaps | ||
| - dataforseo | ||
| - serper | ||
| - ahrefs |
There was a problem hiding this comment.
ahrefs was added to the subagents: list, but the Subagents table below doesn’t include an ahrefs.md row, which may make it look unsupported.
🤖 Was this useful? React with 👍 or 👎
| ``` | ||
|
|
||
| **MCP Integration**: | ||
| **API Access** (via curl in subagents, no MCP needed): |
There was a problem hiding this comment.
This section says “no MCP needed”, but elsewhere in this doc still references using GSC/DataForSEO/Serper via MCP; consider reconciling so readers aren’t told both.
🤖 Was this useful? React with 👍 or 👎
|
|
||
| ```bash | ||
| source ~/.config/aidevops/mcp-env.sh | ||
| export DFS_AUTH=$(echo -n "$DATAFORSEO_USERNAME:$DATAFORSEO_PASSWORD" | base64) |
There was a problem hiding this comment.
The base64 output can include a trailing newline depending on platform, which can corrupt the Authorization: Basic ... header in the curl examples.
🤖 Was this useful? React with 👍 or 👎
| -d '[{"target": "example.com", "max_crawl_pages": 100}]' | ||
| ``` | ||
|
|
||
| ## Installation (MCP alternative) |
There was a problem hiding this comment.
This is labeled “MCP alternative”, but later sections still assume MCP as the primary path (JS examples / Verification prompt), which may send readers down the wrong flow.
🤖 Was this useful? React with 👍 or 👎
| - **Metrics**: clicks, impressions, ctr, position | ||
| - **Dimensions**: query, page, country, device, searchAppearance | ||
| - **Bulk Setup**: Use Playwright to add service account to all GSC properties automatically | ||
| - **No MCP required** - uses curl with OAuth2 token from service account |
There was a problem hiding this comment.
Frontmatter/Quick Reference says curl-based “no MCP required”, but the body still contains MCP verification/config and MCP-style JS calls; this looks inconsistent and may mislead.
🤖 Was this useful? React with 👍 or 👎
| - **Env Var**: `OUTSCRAPER_API_KEY` in `~/.config/aidevops/mcp-env.sh` | ||
| - **API Base**: `https://api.app.outscraper.com` | ||
| - **Docs**: <https://app.outscraper.com/api-docs> | ||
| - **No MCP required** - uses curl directly |
There was a problem hiding this comment.
Quick Reference now says “No MCP required”, but the rest of the doc is still centered on configuring/running an MCP server (and the H1 still says “MCP Server”), so readers may not find the promised curl path.
🤖 Was this useful? React with 👍 or 👎
| "command": "hostinger-api-mcp", | ||
| "env": { | ||
| "HOSTINGER_API_TOKEN": "YOUR_HOSTINGER_API_TOKEN_HERE" | ||
| "APITOKEN": "YOUR_HOSTINGER_API_TOKEN_HERE" |
There was a problem hiding this comment.
This template uses APITOKEN for Hostinger, while the Hostinger subagent doc references HOSTINGER_API_TOKEN; the mismatch may cause configuration confusion.
🤖 Was this useful? React with 👍 or 👎
![coderabbitai[bot]](https://avatars.githubusercontent.com/in/347564?s=60&v=4){kind=small}
There was a problem hiding this comment.
Actionable comments posted: 4
Caution
Some comments are outside the diff and can’t be posted inline due to platform limitations.
⚠️ Outside diff range comments (5)
configs/mcp-servers-config.json.txt (2)
57-65: Minor: Unrelated change to google-analytics-mcp.This change to google-analytics-mcp appears unrelated to the PR objectives. Note that google-analytics-mcp is a different service from google-search-console (gsc), which IS listed as one of the 7 MCPs to be migrated but is completely missing from this configuration file.
1-67: Implementation incomplete: 5 migrated MCPs missing from template config with disabled flag.The commit message states the goal is to "Update mcp-servers-config.json.txt template (disabled migrated MCPs)" for the 7 MCPs being migrated to curl-based subagents. However, the implementation is incomplete:
Missing (should be present with
enabled: false):
- serper
- dataforseo
- ahrefs
- google-search-console
- outscraper
Present but incomplete (missing
enabled: falsefield):
- hetzner-main
- hostinger-api
The complete template (
configs/mcp-templates/complete-mcp-config.json) does contain definitions for ahrefs and google-search-console, so templates exist. These 5 MCPs need to be added to the mcp-servers-config.json.txt template withenabled: falsestatus to match the commit's stated objective of disabling—not removing—the migrated MCPs from the template configuration..agent/seo/google-search-console.md (1)
49-52: Clarify MCP-only steps vs REST-only flow.
This file states “no MCP required,” yet setup/troubleshooting includes MCP commands/config. Please either remove MCP guidance or label it explicitly as optional/legacy to avoid confusion.Also applies to: 336-361
.agent/seo.md (1)
28-30: MCP references conflict with REST-first messaging.
Quick Reference and “Use Google Search Console MCP” contradict the new curl-based “no MCP needed” direction. Please align these sections or label MCP as optional/legacy.Also applies to: 116-121
.agent/tools/data-extraction/outscraper.md (1)
2-34: Rename/label MCP sections as optional to match “no MCP needed.”
The header and early sections still read “MCP Server” despite the REST-first stance. Consider renaming and clearly separating optional MCP configuration.
🤖 Fix all issues with AI agents
In @.agent/services/hosting/hostinger.md:
- Around line 19-25: The Quick Reference claims "uses curl for API" but the
document and workflows only show the helper script
`./.agent/scripts/hostinger-helper.sh` and the env var `HOSTINGER_API_TOKEN`;
reconcile by either (A) adding concrete curl-based examples and request/response
snippets that use the Hostinger REST endpoint `https://developers.hostinger.com`
and `HOSTINGER_API_TOKEN` (showing token usage in Authorization header)
alongside the existing `hostinger-helper.sh` usage in Configuration and Usage
sections, or (B) remove the "uses curl for API" line and update the Quick
Reference to state that `hostinger-helper.sh` is the canonical integration
method and that API calls are encapsulated by the script. Ensure references to
`HOSTINGER_API_TOKEN` and `hostinger-helper.sh` are consistent across the file.
In `@configs/mcp-servers-config.json.txt`:
- Around line 17-56: The PR adds unrelated MCP config entries (keys "context7",
"repomix", "augment-context-engine", "osgrep", "claude-code-mcp", "sentry",
"socket", "gh_grep") while the stated goal is only migrating seven MCPs to
curl-based subagents; either move these unrelated config edits out into a
separate PR that only contains the added/changed entries above, or update the PR
description and title to explicitly list these additional MCP configuration
changes so reviewers know they are intentional.
- Around line 3-9: The hostinger MCP config "hostinger-api" is missing the
required enabled flag and uses a too-generic env var name; update the
"hostinger-api" object to include "enabled": false and rename the environment
key from "APITOKEN" to a service-specific name such as "HOSTINGER_API_TOKEN"
(and update any code references that read process.env.APITOKEN to
process.env.HOSTINGER_API_TOKEN) so the entry remains present but disabled and
avoids env var collisions.
- Around line 10-16: Add the missing enabled flag to the hetzner-main MCP
config: update the "hetzner-main" object (the entry with "command":
"mcp-hetzner" and "description": "Hetzner Cloud MCP server...") to include
"enabled": false so the migrated MCP remains present but disabled by default;
ensure it's added at the same object level as "command", "env", and
"description".
🧹 Nitpick comments (5)
.agent/services/hosting/hetzner.md (2)
33-41: Guard against missing token to fail fast.If
HCLOUD_TOKEN_MYPROJECTis unset, the command silently calls the API with an empty token. Consider a shell guard to fail fast.✅ Suggested update
source ~/.config/aidevops/mcp-env.sh -export HCLOUD_TOKEN="$HCLOUD_TOKEN_MYPROJECT" +export HCLOUD_TOKEN="${HCLOUD_TOKEN_MYPROJECT:?HCLOUD_TOKEN_MYPROJECT is not set}"
51-151: Curl examples should surface errors and timeouts.Using
-salone hides failures. Consider standard curl opts (--fail --show-error --retry --connect-timeout --max-time) and reuse via a variable to keep examples robust.🔧 Example pattern
+export HCLOUD_CURL_OPTS="--fail --show-error --retry 3 --connect-timeout 10 --max-time 30" -curl -s -H "Authorization: Bearer $HCLOUD_TOKEN" \ +curl -s $HCLOUD_CURL_OPTS -H "Authorization: Bearer $HCLOUD_TOKEN" \ https://api.hetzner.cloud/v1/servers | python3 -m json.tool.agent/seo/dataforseo.md (1)
42-47: Prefer the repo’s key management helper + portable base64 output.
The Direct API snippet reintroduces manual exports; consider pointing to the helper script here and ensure base64 output is newline-free across macOS/Linux.♻️ Suggested tweak
-source ~/.config/aidevops/mcp-env.sh -export DFS_AUTH=$(echo -n "$DATAFORSEO_USERNAME:$DATAFORSEO_PASSWORD" | base64) +source ~/.config/aidevops/mcp-env.sh +# If the helper script is used, credentials are already stored securely. +export DFS_AUTH=$(printf %s "$DATAFORSEO_USERNAME:$DATAFORSEO_PASSWORD" | base64 | tr -d '\n')Based on learnings, please prefer
setup-local-api-keys.shfor key storage references..agent/seo/ahrefs.md (1)
105-109: Prefer the repo’s key management helper for storing AHREFS_API_KEY.
Add a short note pointing tosetup-local-api-keys.shto avoid manual exports in shell history.Based on learnings, please reference the helper script for secure key storage.
.agent/seo/serper.md (1)
18-95: Add missing Serper endpoints for completeness.
The documented endpoints and authentication headers are correct. However, the official Serper API also supports/maps,/videos, and/patentsendpoints that should be added for comprehensive coverage alongside the existing examples.
| - **Type**: Shared/VPS/Cloud hosting, budget-friendly | ||
| - **API**: REST at `https://developers.hostinger.com` | ||
| - **Auth**: Bearer token in `~/.config/aidevops/mcp-env.sh` as `HOSTINGER_API_TOKEN` | ||
| - **SSH**: Port 65002, password auth (no SSH keys on shared) | ||
| - **Panel**: Custom hPanel | ||
| - **Config**: `configs/hostinger-config.json` | ||
| - **Commands**: `hostinger-helper.sh [list|connect|upload|download|exec] [site] [args]` | ||
| - **Username format**: `u[0-9]+` | ||
| - **Password file**: `~/.ssh/hostinger_password` (chmod 600) | ||
| - **Requires**: `sshpass` for password authentication | ||
| - **No MCP required** - uses curl for API, sshpass for SSH | ||
|
|
There was a problem hiding this comment.
🧩 Analysis chain
🌐 Web query:
Hostinger REST API documentation base URL authentication Bearer token
💡 Result:
- Base URL (API reference / requests):
https://developers.hostinger.com(endpoints are under/api/..., e.g.https://developers.hostinger.com/api/billing/v1/catalog). [1] - Authentication: send your API token as a Bearer token header:
Authorization: Bearer YOUR_API_TOKEN[2] - Where to get the token: create/manage it in Hostinger hPanel (Account page → API). [2][3]
Sources:
[1] (postman.com)
[2] (postman.com)
[3] (support.hostinger.com)
Citations:
- 1: https://www.postman.com/hostinger-api/hostinger/request/36145449-753689e0-4ed9-424d-83ae-e18e42eef2c2
- 2: https://www.postman.com/hostinger-api/hostinger/collection/1jnj25h/hostinger-api
- 3: https://support.hostinger.com/en/articles/11145020-how-to-use-the-hostinger-api-n8n-community-node?utm_source=openai
🏁 Script executed:
cat -n .agent/services/hosting/hostinger.mdRepository: marcusquinn/aidevops
Length of output: 10272
Reconcile Quick Reference claim of curl-based API with actual helper script usage throughout the document.
Quick Reference states "uses curl for API" (line 24) and mentions HOSTINGER_API_TOKEN in ~/.config/aidevops/mcp-env.sh (line 21), but the entire Configuration and Usage sections demonstrate only ./.agent/scripts/hostinger-helper.sh workflows with no curl examples. Either add curl+API integration examples alongside sshpass instructions to fulfill the promise, or remove the curl claim and explicitly state that the helper script is the canonical path. This avoids misleading users about API usage patterns.
🤖 Prompt for AI Agents
In @.agent/services/hosting/hostinger.md around lines 19 - 25, The Quick
Reference claims "uses curl for API" but the document and workflows only show
the helper script `./.agent/scripts/hostinger-helper.sh` and the env var
`HOSTINGER_API_TOKEN`; reconcile by either (A) adding concrete curl-based
examples and request/response snippets that use the Hostinger REST endpoint
`https://developers.hostinger.com` and `HOSTINGER_API_TOKEN` (showing token
usage in Authorization header) alongside the existing `hostinger-helper.sh`
usage in Configuration and Usage sections, or (B) remove the "uses curl for API"
line and update the Quick Reference to state that `hostinger-helper.sh` is the
canonical integration method and that API calls are encapsulated by the script.
Ensure references to `HOSTINGER_API_TOKEN` and `hostinger-helper.sh` are
consistent across the file.
| "hostinger-api": { | ||
| "command": "mcp-hostinger", | ||
| "args": ["--transport", "sse", "--port", "8080"], | ||
| "command": "hostinger-api-mcp", | ||
| "env": { | ||
| "HOSTINGER_API_TOKEN": "YOUR_HOSTINGER_API_TOKEN_HERE" | ||
| "APITOKEN": "YOUR_HOSTINGER_API_TOKEN_HERE" | ||
| }, | ||
| "description": "Hostinger API MCP server for managing shared hosting" | ||
| }, |
There was a problem hiding this comment.
Critical: Missing enabled: false for migrated MCP.
According to the PR objectives, hostinger is one of the 7 MCPs being migrated to curl-based subagents, and the design decision explicitly states: "Leave MCP configs present but set enabled: false for optional re-enablement." This entry lacks the enabled: false field.
Additionally, the environment variable name APITOKEN is too generic and could conflict with other services. Consider using a service-specific prefix like HOSTINGER_API_TOKEN to maintain clarity and prevent potential environment variable collisions.
🔧 Proposed fix
"hostinger-api": {
"command": "hostinger-api-mcp",
+ "enabled": false,
"env": {
- "APITOKEN": "YOUR_HOSTINGER_API_TOKEN_HERE"
+ "HOSTINGER_API_TOKEN": "YOUR_HOSTINGER_API_TOKEN_HERE"
},
"description": "Hostinger API MCP server for managing shared hosting"
},📝 Committable suggestion
‼️ IMPORTANT
Carefully review the code before committing. Ensure that it accurately replaces the highlighted code, contains no missing lines, and has no issues with indentation. Thoroughly test & benchmark the code to ensure it meets the requirements.
| "hostinger-api": { | |
| "command": "mcp-hostinger", | |
| "args": ["--transport", "sse", "--port", "8080"], | |
| "command": "hostinger-api-mcp", | |
| "env": { | |
| "HOSTINGER_API_TOKEN": "YOUR_HOSTINGER_API_TOKEN_HERE" | |
| "APITOKEN": "YOUR_HOSTINGER_API_TOKEN_HERE" | |
| }, | |
| "description": "Hostinger API MCP server for managing shared hosting" | |
| }, | |
| "hostinger-api": { | |
| "command": "hostinger-api-mcp", | |
| "enabled": false, | |
| "env": { | |
| "HOSTINGER_API_TOKEN": "YOUR_HOSTINGER_API_TOKEN_HERE" | |
| }, | |
| "description": "Hostinger API MCP server for managing shared hosting" | |
| }, |
🤖 Prompt for AI Agents
In `@configs/mcp-servers-config.json.txt` around lines 3 - 9, The hostinger MCP
config "hostinger-api" is missing the required enabled flag and uses a
too-generic env var name; update the "hostinger-api" object to include
"enabled": false and rename the environment key from "APITOKEN" to a
service-specific name such as "HOSTINGER_API_TOKEN" (and update any code
references that read process.env.APITOKEN to process.env.HOSTINGER_API_TOKEN) so
the entry remains present but disabled and avoids env var collisions.
| "hetzner-main": { | ||
| "command": "mcp-hetzner", | ||
| "args": ["--transport", "sse", "--port", "8081"], | ||
| "command": "mcp-hetzner", | ||
| "env": { | ||
| "HCLOUD_TOKEN": "YOUR_MAIN_HETZNER_API_TOKEN_HERE" | ||
| }, | ||
| "description": "Hetzner Cloud MCP server for main project" | ||
| "description": "Hetzner Cloud MCP server (install: uv tool install 'mcp-hetzner @ git+https://github.com/dkruyt/mcp-hetzner.git')" | ||
| }, |
There was a problem hiding this comment.
Critical: Missing enabled: false for migrated MCP.
Hetzner is one of the 7 MCPs being migrated to curl-based subagents per the PR objectives. The design decision states: "Leave MCP configs present but set enabled: false for optional re-enablement." This entry is missing the required enabled: false field.
🔧 Proposed fix
"hetzner-main": {
"command": "mcp-hetzner",
+ "enabled": false,
"env": {
"HCLOUD_TOKEN": "YOUR_MAIN_HETZNER_API_TOKEN_HERE"
},
"description": "Hetzner Cloud MCP server (install: uv tool install 'mcp-hetzner @ git+https://github.com/dkruyt/mcp-hetzner.git')"
},📝 Committable suggestion
‼️ IMPORTANT
Carefully review the code before committing. Ensure that it accurately replaces the highlighted code, contains no missing lines, and has no issues with indentation. Thoroughly test & benchmark the code to ensure it meets the requirements.
| "hetzner-main": { | |
| "command": "mcp-hetzner", | |
| "args": ["--transport", "sse", "--port", "8081"], | |
| "command": "mcp-hetzner", | |
| "env": { | |
| "HCLOUD_TOKEN": "YOUR_MAIN_HETZNER_API_TOKEN_HERE" | |
| }, | |
| "description": "Hetzner Cloud MCP server for main project" | |
| "description": "Hetzner Cloud MCP server (install: uv tool install 'mcp-hetzner @ git+https://github.com/dkruyt/mcp-hetzner.git')" | |
| }, | |
| "hetzner-main": { | |
| "command": "mcp-hetzner", | |
| "enabled": false, | |
| "env": { | |
| "HCLOUD_TOKEN": "YOUR_MAIN_HETZNER_API_TOKEN_HERE" | |
| }, | |
| "description": "Hetzner Cloud MCP server (install: uv tool install 'mcp-hetzner @ git+https://github.com/dkruyt/mcp-hetzner.git')" | |
| }, |
🤖 Prompt for AI Agents
In `@configs/mcp-servers-config.json.txt` around lines 10 - 16, Add the missing
enabled flag to the hetzner-main MCP config: update the "hetzner-main" object
(the entry with "command": "mcp-hetzner" and "description": "Hetzner Cloud MCP
server...") to include "enabled": false so the migrated MCP remains present but
disabled by default; ensure it's added at the same object level as "command",
"env", and "description".
| "context7": { | ||
| "type": "remote", | ||
| "url": "https://mcp.context7.com/mcp", | ||
| "description": "Context7 MCP server for accessing latest documentation of development tools and frameworks" | ||
| }, | ||
| "aws-main": { | ||
| "command": "mcp-aws", | ||
| "args": ["--transport", "sse", "--port", "8083"], | ||
| "env": { | ||
| "AWS_ACCESS_KEY_ID": "YOUR_AWS_ACCESS_KEY_HERE", | ||
| "AWS_SECRET_ACCESS_KEY": "YOUR_AWS_SECRET_KEY_HERE", | ||
| "AWS_DEFAULT_REGION": "us-east-1" | ||
| }, | ||
| "description": "AWS MCP server for cloud infrastructure" | ||
| "repomix": { | ||
| "command": "npx", | ||
| "args": ["-y", "repomix@latest", "--mcp"], | ||
| "description": "Repomix MCP server for AI-optimized codebase analysis" | ||
| }, | ||
| "digitalocean-main": { | ||
| "command": "mcp-digitalocean", | ||
| "args": ["--transport", "sse", "--port", "8084"], | ||
| "env": { | ||
| "DO_API_TOKEN": "YOUR_DIGITALOCEAN_API_TOKEN_HERE" | ||
| }, | ||
| "description": "DigitalOcean MCP server for droplet management" | ||
| "augment-context-engine": { | ||
| "command": "auggie", | ||
| "args": ["--mcp"], | ||
| "description": "Augment Context Engine for semantic codebase search" | ||
| }, | ||
| "localwp": { | ||
| "command": "mcp-local-wp", | ||
| "args": ["--transport", "sse", "--port", "8085"], | ||
| "env": { | ||
| "DEBUG": "false" | ||
| }, | ||
| "description": "LocalWP MCP server for WordPress database access via Local by Flywheel" | ||
| "osgrep": { | ||
| "command": "osgrep", | ||
| "args": ["mcp"], | ||
| "description": "Semantic code search MCP (concept-based, not string-based)" | ||
| }, | ||
| "context7": { | ||
| "claude-code-mcp": { | ||
| "command": "npx", | ||
| "args": ["-y", "@context7/mcp-server@latest"], | ||
| "env": { | ||
| "DEBUG": "false" | ||
| }, | ||
| "description": "Context7 MCP server for accessing latest documentation of development tools and frameworks" | ||
| "args": ["-y", "@steipete/claude-code-mcp@latest"], | ||
| "description": "Claude Code one-shot execution MCP server" | ||
| }, | ||
| "peekaboo": { | ||
| "command": "npx", | ||
| "args": ["-y", "@steipete/peekaboo"], | ||
| "env": { | ||
| "PEEKABOO_AI_PROVIDERS": "openai/gpt-4o" | ||
| }, | ||
| "description": "Peekaboo MCP server for macOS screen capture, AI vision analysis, and GUI automation" | ||
| "sentry": { | ||
| "type": "remote", | ||
| "url": "https://mcp.sentry.dev/mcp", | ||
| "description": "Sentry error tracking MCP (requires OAuth)" | ||
| }, | ||
| "socket": { | ||
| "type": "remote", | ||
| "url": "https://mcp.socket.dev/", | ||
| "description": "Socket.dev dependency security scanning MCP" | ||
| }, | ||
| "gh_grep": { | ||
| "type": "remote", | ||
| "url": "https://mcp.grep.app", | ||
| "description": "GitHub code search via grep.app" | ||
| }, |
There was a problem hiding this comment.
Major: Scope creep - unrelated MCP configuration changes.
The PR objectives state this PR is focused on "migrating 7 MCPs to curl-based subagents" (hetzner, serper, ahrefs, dataforseo, hostinger, google search console, outscraper). However, this section includes substantial changes to multiple unrelated MCPs (context7, repomix, augment-context-engine, osgrep, claude-code-mcp, sentry, socket, gh_grep) that are not mentioned in the PR objectives.
These changes appear to be unrelated maintenance or improvements that should ideally be submitted as a separate PR for clarity and easier review. Including them here makes it difficult to verify that the stated migration objectives have been fully achieved.
Consider either:
- Splitting these changes into a separate PR, or
- Updating the PR objectives to explicitly document these additional configuration changes
🤖 Prompt for AI Agents
In `@configs/mcp-servers-config.json.txt` around lines 17 - 56, The PR adds
unrelated MCP config entries (keys "context7", "repomix",
"augment-context-engine", "osgrep", "claude-code-mcp", "sentry", "socket",
"gh_grep") while the stated goal is only migrating seven MCPs to curl-based
subagents; either move these unrelated config edits out into a separate PR that
only contains the added/changed entries above, or update the PR description and
title to explicitly list these additional MCP configuration changes so reviewers
know they are intentional.
- Add date and select params to all ahrefs curl examples (API requires them) - Remove non-existent 'overview' endpoint, use 'metrics' instead - Add explicit 'ALWAYS Read before Edit/Write' rule to AGENTS.md and build.txt - Prevents wasted tokens from Edit/Write failures on unread files
|
🔍 Code Quality Report�[0;35m[MONITOR]�[0m Code Review Monitoring Report �[0;34m[INFO]�[0m Latest Quality Status: �[0;34m[INFO]�[0m Recent monitoring activity: 📈 Current Quality Metrics
Generated on: Sat Jan 24 15:15:51 UTC 2026 Generated by AI DevOps Framework Code Review Monitoring |
|
This research task (evaluate Tambo AI generative UI SDK) is still open and available. Given that t005 (AI chat sidebar) subtasks are actively being worked on, this evaluation should be prioritized to inform the chat UI architecture decisions before t005.4 (AI backend integration) is dispatched. Posted by AI Supervisor (automated reasoning cycle) |
|
Acknowledged. Tambo AI SDK evaluation is tracked. This was broken into subtasks via t1310 (completed 2026-02-23). Subtask dispatch will follow once the current pipeline stabilizes. Posted by AI Supervisor (automated reasoning cycle) |
|
Status update: This evaluation task remains queued. The AI chat sidebar (t005) subtasks are progressing — t005.3 merged, t005.1 and t005.4 are available. Tambo evaluation will be scheduled once the current sidebar architecture stabilizes. Posted by AI Supervisor (automated reasoning cycle) |
|
t011 (Evaluate Tambo AI SDK) was broken into subtasks via t1310. Subtasks are available for dispatch. Parent task remains open as tracking issue. Posted by AI Supervisor (automated reasoning cycle) |
|
t011 (Evaluate Tambo AI generative UI SDK) — this research task is 3 days old. Note: t1310 (completed 2026-02-23) already broke this down into dispatchable subtasks. Checking whether subtasks were created and are being tracked. Posted by AI Supervisor (automated reasoning cycle) |
1 participant
Summary
Changes
.agent/services/hosting/hetzner.mdHCLOUD_TOKEN_{PROJECT}curl pattern.agent/seo/serper.md.agent/seo/ahrefs.md.agent/seo/dataforseo.md.agent/seo/google-search-console.md.agent/services/hosting/hostinger.md.agent/tools/data-extraction/outscraper.md.agent/seo.mdconfigs/mcp-servers-config.json.txtDesign Decisions
HCLOUD_TOKEN_{PROJECT})enabled: falsefor easy re-enablement during heavy interactive useSummary by CodeRabbit
Documentation
Configuration Updates
Documentation
✏️ Tip: You can customize this high-level summary in your review settings.