Skip to content

Since.dev llms.txt changelog

io.github.since-dev/mcp

3 changes recorded. MCP Harbor re-reads this listing from the official registry every six hours and asks the live server what it exposes, and keeps what changed between one look and the next.

  • llms.txt +15 −81

    https://since.dev/llms.txt

    − > Grant by Since.dev is the COO for your software operations: an AI agent who leads a team of specialists. They watch your GitHub Actions CI, cloud and AI costs, runtime health, security findings and dependency changes across your projects, then bring you what matters with the evidence. Ask in Slack, the dashboard or your agent's MCP client. Check-ins only read, except Vera's tests of your own agents. Supported repairs arrive as pull requests for your review and nothing merges or deploys. Since − Grant leads a preset team of specialists. Their check-ins and consults only read authorized, supported sources, such as GitHub, AWS, AI provider billing and uploaded documents, except that Vera sends test prompts to agent endpoints the owner confirms they own. Coverage gaps and pauses stay visible. Web, Slack and MCP reach the same Grant and team under the same workspace permissions, and share findings and saved decisions.− Since.dev also opens draft PRs for eligible compatibility repairs under separate repository permission and repair settings; where GitHub offers no draft PRs, the repair opens as a regular PR titled [Needs review]. CI runs in your repository and people decide whether to merge. World State verification returns valid, changed or unknown; unknown does not mean unchanged.− - [Meet Grant](https://since.dev/docs#grant): Grant, your COO, his team of AI agents and what their check-ins may read.− - [Grant onboarding](https://since.dev/docs#grant-onboarding): Connect GitHub or AWS, choose project scope, then start Grant. Bring your own AI provider if you want to.− - [Grant's team](https://since.dev/docs#grant-team): Each specialist's role, sources and check-ins.− - [Conversations and follow-through](https://since.dev/docs#grant-conversation): Consults, background assignments, findings and saved decisions.− - [Grant delivery](https://since.dev/docs#grant-delivery): Slack and signed webhook notices.− - [Quickstart](https://since.dev/docs#quickstart): Create a key, watch the dependencies you use, then read what changed.− - [Concepts](https://since.dev/docs#concepts): Watches, sets, changes and classifications.− - [World State webhook delivery](https://since.dev/docs#webhooks): Signed, at-least-once; dedupe on idempotencyKey.− - [GitHub repositories and repairs](https://since.dev/docs#repairs): Monitor only and Automatic draft PRs.− - [Grant MCP tools](https://since.dev/docs#mcp-grant): Team status, saved reads, questions, stop and decisions with explicit project and action permissions; Since AI credits apply to questions unless the workspace uses its own AI provider.− Practical software operations guides for founders and CTOs of small teams.− - [Software operations guides for startups](https://since.dev/guides): Software operations guides for founders and CTOs: the startup COO role, DevOps, AWS and AI costs, security, dependencies, AI agents, MCP and a weekly review.− - [DevOps for startups without a DevOps team](https://since.dev/guides/devops-for-startups): Set up four things once, then check five every week. A practical DevOps checklist for startups on GitHub and AWS that have no DevOps engineer yet.− - [What does a COO do at a startup?](https://since.dev/guides/startup-coo): What a startup COO owns, when companies add one, how the role differs from CTO and VP of Engineering, and who runs software operations before then.− - [Weekly operations review: a template for small software teams](https://since.dev/guides/weekly-operations-review): A 30-minute weekly operations review for small software teams: the agenda, the checks for CI, AWS, security and dependencies, and a decision log to copy.− - [When to hire your first DevOps engineer](https://since.dev/guides/first-devops-engineer): The signs you need a DevOps engineer, what the first one should own, and how a full-time hire compares with a contractor, a managed platform or software.− - [AWS cost monitoring for startups: a 15-minute weekly review](https://since.dev/guides/aws-cost-monitoring): Turn on free AWS cost alerts, read Cost Explorer by service each week and check the common leaks: NAT gateways, idle volumes, log retention and idle RDS.− - [How to triage Dependabot, code scanning and Security Hub alerts](https://since.dev/guides/security-alert-triage): A 20-minute weekly routine to triage Dependabot, code scanning and AWS Security Hub alerts: known exploitation first, then exposure, then severity.− - [AI credits: what one costs and how to budget](https://since.dev/guides/ai-credits): AI credits are each vendor's own unit for AI use. Learn how credits differ from tokens, turn one into dollars and budget a month before you buy.− - [Bring your own key (BYOK) vs hosted AI: which to choose](https://since.dev/guides/byok-vs-hosted-ai): Bring your own key (BYOK) means an AI tool runs on your own OpenAI or Anthropic LLM API key. When it beats hosted AI, and how to set it up safely.− - [How to track OpenAI and Anthropic API usage and costs](https://since.dev/guides/llm-api-cost-monitoring): Track OpenAI and Anthropic API usage and costs with what the providers already give you: usage pages, cost APIs, spend limits and a weekly LLM cost review.− - [How to connect Claude or ChatGPT to an MCP server](https://since.dev/guides/mcp-for-operations): What an MCP server is, how to add a remote one to Claude, ChatGPT or Cursor, a read-only GitHub example, and OAuth vs API keys.− - [AI agents for DevOps: what to hand off and what to keep](https://since.dev/guides/ai-agents-for-devops): Which DevOps jobs a small team can hand to AI agents today, which ones a person should keep, and how to trial an agent in a week without risking production.− - grant_team_status− - grant_read− - grant_question− - grant_stop− - grant_record_decision− - grant_alert− - get_usage− - what_changed_since− - acknowledge_changes− - verify_facts− - subscribe− - list_sets− - attach_to_set− - check_coverage+ > Since.dev tracks every package, AI model, API and runtime as it changes, finds the exact lines a change breaks in your code and opens the pull request that fixes it. Use it from the dashboard, the HTTP API or your agent's MCP client. Nothing merges or deploys: people review every pull request.+ Since.dev tracks every package, API and AI model, and analyzes the repositories you scan or connect to find the code a change hits. On Team and up it opens fix pull requests, ready for review. CI runs in your repository and people decide whether to merge. Nothing merges or deploys.+ - [Quickstart](https://since.dev/docs#quickstart): Create a key, read what changed, then check your code against it.+ - [Concepts](https://since.dev/docs#concepts): Surfaces, changes, kinds, severity and what is not settled.+ - [Webhooks](https://since.dev/docs#webhooks): Signed, at-least-once; dedupe on idempotencyKey.+ - [GitHub repositories and fix pull requests](https://since.dev/docs#repairs): Impacts only on Free, fix PRs on Team and up.+ Practical guides for developers on tracking and fixing what their dependencies change.+ - [Guides for tracking dependency changes](https://since.dev/guides): Guides for developers on catching breaking changes, deprecations and end-of-life dates in the packages, APIs and runtimes a repository depends on.+ - whats_changed+ - check_code+ - POST /v1/check+ - POST /v1/webhooks+ - GET /v1/repos+ - [Retirement calendar](https://since.dev/calendar): Dated retirements and end of life dates for AI models, provider APIs and runtimes.+ - [Calendar API](https://api.since.dev/v1/calendar): The same dates as JSON, free and without a key. Add .ics for an iCal feed or .rss for RSS.

    Showing the first 40 of each.

  • llms.txt +11 −6

    https://since.dev/llms.txt

    − > Grant by Since.dev is the COO for your software operations: an AI agent who leads a team of specialists. They watch your GitHub Actions CI, cloud and AI costs, runtime health, security findings and dependency changes across your projects, then bring you what matters with the evidence. Ask in Slack, the dashboard or your agent's MCP client. Check-ins only read, except Vera's tests of your own agents. Supported repairs arrive as pull requests for your review and nothing merges or deploys. Paid p− - [Grant onboarding](https://since.dev/docs#grant-onboarding): Connect GitHub or AWS, choose project scope and a model, then start Grant.− - [Grant MCP tools](https://since.dev/docs#mcp-grant): Team status, saved reads, questions, stop and decisions with explicit project and action permissions; workspace allowances apply to questions.− - [Software operations guides for startups](https://since.dev/guides): Software operations guides for founders and CTOs: the startup COO role, a weekly operations review, DevOps, AWS costs, security and dependencies.− - Maya: CI and delivery. Checks in on workflow runs and checks.− - Owen: Configuration and lifecycle. Checks in on pushes.+ > Grant by Since.dev is the COO for your software operations: an AI agent who leads a team of specialists. They watch your GitHub Actions CI, cloud and AI costs, runtime health, security findings and dependency changes across your projects, then bring you what matters with the evidence. Ask in Slack, the dashboard or your agent's MCP client. Check-ins only read, except Vera's tests of your own agents. Supported repairs arrive as pull requests for your review and nothing merges or deploys. Since + - [Grant onboarding](https://since.dev/docs#grant-onboarding): Connect GitHub or AWS, choose project scope, then start Grant. Bring your own AI provider if you want to.+ - [Grant MCP tools](https://since.dev/docs#mcp-grant): Team status, saved reads, questions, stop and decisions with explicit project and action permissions; Since AI credits apply to questions unless the workspace uses its own AI provider.+ - [Software operations guides for startups](https://since.dev/guides): Software operations guides for founders and CTOs: the startup COO role, DevOps, AWS and AI costs, security, dependencies, AI agents, MCP and a weekly review.+ - [AI credits: what one costs and how to budget](https://since.dev/guides/ai-credits): AI credits are each vendor's own unit for AI use. Learn how credits differ from tokens, turn one into dollars and budget a month before you buy.+ - [Bring your own key (BYOK) vs hosted AI: which to choose](https://since.dev/guides/byok-vs-hosted-ai): Bring your own key (BYOK) means an AI tool runs on your own OpenAI or Anthropic LLM API key. When it beats hosted AI, and how to set it up safely.+ - [How to track OpenAI and Anthropic API usage and costs](https://since.dev/guides/llm-api-cost-monitoring): Track OpenAI and Anthropic API usage and costs with what the providers already give you: usage pages, cost APIs, spend limits and a weekly LLM cost review.+ - [How to connect Claude or ChatGPT to an MCP server](https://since.dev/guides/mcp-for-operations): What an MCP server is, how to add a remote one to Claude, ChatGPT or Cursor, a read-only GitHub example, and OAuth vs API keys.+ - [AI agents for DevOps: what to hand off and what to keep](https://since.dev/guides/ai-agents-for-devops): Which DevOps jobs a small team can hand to AI agents today, which ones a person should keep, and how to trial an agent in a week without risking production.+ - Maya: CI and delivery. Checks in on default-branch workflow runs and checks.+ - Owen: Configuration and lifecycle. Checks in on default-branch pushes.

  • llms.txt +42 −3

    https://since.dev/llms.txt

    − > Since.dev helps software recover from upstream change. It watches APIs, packages, SDKs and models, finds affected code, and prepares supported repairs as draft PRs. Agents can also watch changes and check current facts through MCP, HTTP and signed webhooks.− Since.dev opens draft PRs only for supported repairs. CI runs in your repository and people decide whether to merge. Verification returns valid, changed or unknown; unknown does not mean unchanged.− - [Webhook delivery](https://since.dev/docs#webhooks): Signed, at-least-once; dedupe on idempotencyKey.+ > Grant by Since.dev is the COO for your software operations: an AI agent who leads a team of specialists. They watch your GitHub Actions CI, cloud and AI costs, runtime health, security findings and dependency changes across your projects, then bring you what matters with the evidence. Ask in Slack, the dashboard or your agent's MCP client. Check-ins only read, except Vera's tests of your own agents. Supported repairs arrive as pull requests for your review and nothing merges or deploys. Paid p+ Grant leads a preset team of specialists. Their check-ins and consults only read authorized, supported sources, such as GitHub, AWS, AI provider billing and uploaded documents, except that Vera sends test prompts to agent endpoints the owner confirms they own. Coverage gaps and pauses stay visible. Web, Slack and MCP reach the same Grant and team under the same workspace permissions, and share findings and saved decisions.+ Since.dev also opens draft PRs for eligible compatibility repairs under separate repository permission and repair settings; where GitHub offers no draft PRs, the repair opens as a regular PR titled [Needs review]. CI runs in your repository and people decide whether to merge. World State verification returns valid, changed or unknown; unknown does not mean unchanged.+ - [Meet Grant](https://since.dev/docs#grant): Grant, your COO, his team of AI agents and what their check-ins may read.+ - [Grant onboarding](https://since.dev/docs#grant-onboarding): Connect GitHub or AWS, choose project scope and a model, then start Grant.+ - [Grant's team](https://since.dev/docs#grant-team): Each specialist's role, sources and check-ins.+ - [Conversations and follow-through](https://since.dev/docs#grant-conversation): Consults, background assignments, findings and saved decisions.+ - [Grant delivery](https://since.dev/docs#grant-delivery): Slack and signed webhook notices.+ - [World State webhook delivery](https://since.dev/docs#webhooks): Signed, at-least-once; dedupe on idempotencyKey.+ - [Grant MCP tools](https://since.dev/docs#mcp-grant): Team status, saved reads, questions, stop and decisions with explicit project and action permissions; workspace allowances apply to questions.+ ## Guides+ Practical software operations guides for founders and CTOs of small teams.+ - [Software operations guides for startups](https://since.dev/guides): Software operations guides for founders and CTOs: the startup COO role, a weekly operations review, DevOps, AWS costs, security and dependencies.+ - [DevOps for startups without a DevOps team](https://since.dev/guides/devops-for-startups): Set up four things once, then check five every week. A practical DevOps checklist for startups on GitHub and AWS that have no DevOps engineer yet.+ - [What does a COO do at a startup?](https://since.dev/guides/startup-coo): What a startup COO owns, when companies add one, how the role differs from CTO and VP of Engineering, and who runs software operations before then.+ - [Weekly operations review: a template for small software teams](https://since.dev/guides/weekly-operations-review): A 30-minute weekly operations review for small software teams: the agenda, the checks for CI, AWS, security and dependencies, and a decision log to copy.+ - [When to hire your first DevOps engineer](https://since.dev/guides/first-devops-engineer): The signs you need a DevOps engineer, what the first one should own, and how a full-time hire compares with a contractor, a managed platform or software.+ - [AWS cost monitoring for startups: a 15-minute weekly review](https://since.dev/guides/aws-cost-monitoring): Turn on free AWS cost alerts, read Cost Explorer by service each week and check the common leaks: NAT gateways, idle volumes, log retention and idle RDS.+ - [How to triage Dependabot, code scanning and Security Hub alerts](https://since.dev/guides/security-alert-triage): A 20-minute weekly routine to triage Dependabot, code scanning and AWS Security Hub alerts: known exploitation first, then exposure, then severity.+ - [How to track breaking changes and deprecations in your dependencies](https://since.dev/guides/dependency-breaking-changes): Dependabot flags vulnerable and outdated versions, not announced breaking changes. Where deprecation and end-of-life notices appear, and how to catch them.+ - grant_team_status+ - grant_read+ - grant_question+ - grant_stop+ - grant_record_decision+ - grant_alert+ - inspect_repository+ - list_watches+ - inspect_watch+ ## Grant's team+ Grant and every specialist are AI agents, not people.+ - Grant: COO. Answers you, consults the team, follows your decisions.+ - David: Compatibility and repairs. Reports repairs; supported fixes arrive as draft PRs.+ - Maya: CI and delivery. Checks in on workflow runs and checks.+ - Liz: Costs. Estimates are not realized savings.+ - Noah: Runtime. Missing metrics are unknown, not healthy.+ - Priya: Security. Never starts scans; checks in on new alerts.+ - Owen: Configuration and lifecycle. Checks in on pushes.+ - Iris: World State. Sets up watches and runs fact checks on request only.+ - Harper: Chief of Staff. Writes reports, one-pagers, decks and the Monday brief from the team’s briefs.