Source: https://internal-agents.com/agents/sentry-junior # Sentry — Junior An open-source Slack agent built at Sentry that acts like an intern; takes tasks, retrieves context across many company systems, and is steered and reviewed by humans. Its CEO argues one general-purpose agent beat several vendor-specific bots. - Approach type: Task agent - Deployment stage: Deployed - Autonomy: Human in loop - Evidence strength: Detailed primary - Status: Open sourced - First reported year: 2026 - Work: Coding, Code review, Support, On-call - Interfaces: Slack, Web, Github - Invocation: Event driven, Interactive - Entry reviewed: 2026-09-09 ## Where people stay involved - **assigned task → human-steered and reviewed output** — Continuous steering · Level 2 ## Overview ### Summary An open-source Slack agent built at Sentry that acts like an intern; takes tasks, retrieves context across many company systems, and is steered and reviewed by humans. Its CEO argues one general-purpose agent beat several vendor-specific bots. Fact · Reported · High confidence · `sentry-junior--summary` Confidence reason: A linked first-party source states the claim. Evidence: - Supports · [1] [Building an Intern (Junior at Sentry)](https://cra.mr/building-an-intern/) · [Preserved copy](https://github.com/steel-experiments/internal-agents-map/blob/main/archive/sources/sentry-junior-source-1/content.md) - Supports · [2] [getsentry/junior source repository](https://github.com/getsentry/junior) · [Preserved copy](https://github.com/steel-experiments/internal-agents-map/blob/main/archive/sources/sentry-junior-source-2/content.md) - Contextualizes · [3] [Sentry Labs](https://labs.sentry.dev/) · [Preserved copy](https://github.com/steel-experiments/internal-agents-map/blob/main/archive/sources/sentry-junior-source-3/content.md) ## How it works ### Supporting component searchMcpTools loads tools on demand instead of dumping every schema into the prompt Fact · Reported · High confidence · `sentry-junior--primitives-0` Confidence reason: A linked first-party source states the claim. Evidence: - Supports · [1] [Building an Intern (Junior at Sentry)](https://cra.mr/building-an-intern/) · [Preserved copy](https://github.com/steel-experiments/internal-agents-map/blob/main/archive/sources/sentry-junior-source-1/content.md) ### Supporting component Credentials injected host-side; the sandbox/model never touches a secret Fact · Reported · High confidence · `sentry-junior--primitives-1` Confidence reason: A linked first-party source states the claim. Evidence: - Supports · [1] [Building an Intern (Junior at Sentry)](https://cra.mr/building-an-intern/) · [Preserved copy](https://github.com/steel-experiments/internal-agents-map/blob/main/archive/sources/sentry-junior-source-1/content.md) ### Supporting component Survives serverless timeouts via a queue + claim model Fact · Reported · High confidence · `sentry-junior--primitives-2` Confidence reason: A linked first-party source states the claim. Evidence: - Supports · [1] [Building an Intern (Junior at Sentry)](https://cra.mr/building-an-intern/) · [Preserved copy](https://github.com/steel-experiments/internal-agents-map/blob/main/archive/sources/sentry-junior-source-1/content.md) ## Supervision evidence ### Operating model assessment Level 2 for assigned task → human-steered and reviewed output; human attention boundary: continuous-steering. Inference · Catalog judgment · High confidence · `sentry-junior--operating-models-0` Confidence reason: The source explicitly describes humans steering and reviewing the agent throughout its work. Qualifications: - Observation date: 2026 Evidence: - Supports · [1] [Building an Intern (Junior at Sentry)](https://cra.mr/building-an-intern/) · [Preserved copy](https://github.com/steel-experiments/internal-agents-map/blob/main/archive/sources/sentry-junior-source-1/content.md) ## Implementation details ### Sandbox Vercel serverless functions; Vercel agent-browser sandbox with an on-path proxy for traffic interception; ephemeral containers Fact · Reported · High confidence · `sentry-junior--architecture-sandbox` Confidence reason: A linked first-party source states the claim. Evidence: - Supports · [1] [Building an Intern (Junior at Sentry)](https://cra.mr/building-an-intern/) · [Preserved copy](https://github.com/steel-experiments/internal-agents-map/blob/main/archive/sources/sentry-junior-source-1/content.md) ### Harness Custom harness on Pi's SDK; a task broker over Vercel Queues with an inbox -> worker-claim -> interrupt/resume pattern to survive serverless timeouts; 'skills-as-runbooks' Fact · Reported · High confidence · `sentry-junior--architecture-harness` Confidence reason: A linked first-party source states the claim. Evidence: - Supports · [1] [Building an Intern (Junior at Sentry)](https://cra.mr/building-an-intern/) · [Preserved copy](https://github.com/steel-experiments/internal-agents-map/blob/main/archive/sources/sentry-junior-source-1/content.md) ### Model Claude Sonnet (faster); swappable (Opus as a more expensive option) Fact · Reported · High confidence · `sentry-junior--architecture-model` Confidence reason: A linked first-party source states the claim. Evidence: - Supports · [1] [Building an Intern (Junior at Sentry)](https://cra.mr/building-an-intern/) · [Preserved copy](https://github.com/steel-experiments/internal-agents-map/blob/main/archive/sources/sentry-junior-source-1/content.md) ### Interfaces slack, web, github Fact · Reported · High confidence · `sentry-junior--architecture-interfaces` Confidence reason: A linked first-party source states the claim. Evidence: - Supports · [1] [Building an Intern (Junior at Sentry)](https://cra.mr/building-an-intern/) · [Preserved copy](https://github.com/steel-experiments/internal-agents-map/blob/main/archive/sources/sentry-junior-source-1/content.md) ### Tool access Progressive discovery via MCP; by default Junior connects to no provider until the agent requests a tool lookup; plugins connect Sentry, GitHub, Linear, Notion Fact · Reported · High confidence · `sentry-junior--architecture-tool-access` Confidence reason: A linked first-party source states the claim. Evidence: - Supports · [1] [Building an Intern (Junior at Sentry)](https://cra.mr/building-an-intern/) · [Preserved copy](https://github.com/steel-experiments/internal-agents-map/blob/main/archive/sources/sentry-junior-source-1/content.md) ### Knowledge Conversation transcripts persisted in Redis; repo search to trace code paths; skill docs (TELEMETRY.md, SOUL.md) Fact · Reported · High confidence · `sentry-junior--architecture-knowledge` Confidence reason: A linked first-party source states the claim. Evidence: - Supports · [1] [Building an Intern (Junior at Sentry)](https://cra.mr/building-an-intern/) · [Preserved copy](https://github.com/steel-experiments/internal-agents-map/blob/main/archive/sources/sentry-junior-source-1/content.md) ### Credentials On-path proxy injection; the model never sees the token because it is not in the sandbox; plugins declare OAuth flows and credential domains; GitHub distinguishes read vs. write Fact · Reported · High confidence · `sentry-junior--architecture-credentials` Confidence reason: A linked first-party source states the claim. Evidence: - Supports · [1] [Building an Intern (Junior at Sentry)](https://cra.mr/building-an-intern/) · [Preserved copy](https://github.com/steel-experiments/internal-agents-map/blob/main/archive/sources/sentry-junior-source-1/content.md) ### Context management Incremental transcript updates in Redis; resource subscriptions to GitHub PR events for follow-ups Fact · Reported · High confidence · `sentry-junior--architecture-context-mgmt` Confidence reason: A linked first-party source states the claim. Evidence: - Supports · [1] [Building an Intern (Junior at Sentry)](https://cra.mr/building-an-intern/) · [Preserved copy](https://github.com/steel-experiments/internal-agents-map/blob/main/archive/sources/sentry-junior-source-1/content.md) ## Reported metrics ### Headline claim Open-source (Apache-2.0) Slack agent (~100k lines of TS) used internally at Sentry Metric · Reported · High confidence · `sentry-junior--headline-metric` Confidence reason: A linked first-party source states the claim. Qualifications: - The source does not report the denominator of this figure. - Reported by: Sentry - Scope: Junior codebase size and license in the author report Evidence: - Supports · [1] [Building an Intern (Junior at Sentry)](https://cra.mr/building-an-intern/) · [Preserved copy](https://github.com/steel-experiments/internal-agents-map/blob/main/archive/sources/sentry-junior-source-1/content.md) · Preserved content.md, lines 10–24, 34–60 - Supports · [4] [getsentry/junior Apache 2.0 license](https://github.com/getsentry/junior/blob/main/LICENSE?plain=1) · [Preserved copy](https://github.com/steel-experiments/internal-agents-map/blob/main/archive/sources/sentry-junior-source-4/content.md) · Preserved content.md, lines 412–414 ### Key observation Around 100,000 lines of TypeScript excluding tests, evals, docs, and lockfiles Metric · Reported · High confidence · `sentry-junior--key-metrics-0` Confidence reason: A linked first-party source states the claim. Qualifications: - The source does not report the denominator of this figure. - Reported by: Sentry - Scope: TypeScript lines in Junior excluding tests, evals, documentation, and lockfiles Evidence: - Supports · [1] [Building an Intern (Junior at Sentry)](https://cra.mr/building-an-intern/) · [Preserved copy](https://github.com/steel-experiments/internal-agents-map/blob/main/archive/sources/sentry-junior-source-1/content.md) · Preserved content.md, lines 22 ### Key observation 4 months from start to writeup Metric · Reported · High confidence · `sentry-junior--key-metrics-1` Confidence reason: A linked first-party source states the claim. Qualifications: - The source does not report the denominator of this figure. - Reported by: Sentry - Scope: Author-reported development and iteration time before the writeup Evidence: - Supports · [1] [Building an Intern (Junior at Sentry)](https://cra.mr/building-an-intern/) · [Preserved copy](https://github.com/steel-experiments/internal-agents-map/blob/main/archive/sources/sentry-junior-source-1/content.md) · Preserved content.md, lines 10–22 ## Lessons and interpretation ### Lesson One general-purpose agent connected to many company systems beats several vendor-specific bots Inference · Catalog judgment · Medium confidence · `sentry-junior--lessons-learned-0` Confidence reason: The catalog derives this observation from the linked sources. Evidence: - Supports · [1] [Building an Intern (Junior at Sentry)](https://cra.mr/building-an-intern/) · [Preserved copy](https://github.com/steel-experiments/internal-agents-map/blob/main/archive/sources/sentry-junior-source-1/content.md) ### Lesson Skills-as-runbooks encode operational knowledge the agent can follow Inference · Catalog judgment · Medium confidence · `sentry-junior--lessons-learned-1` Confidence reason: The catalog derives this observation from the linked sources. Evidence: - Supports · [1] [Building an Intern (Junior at Sentry)](https://cra.mr/building-an-intern/) · [Preserved copy](https://github.com/steel-experiments/internal-agents-map/blob/main/archive/sources/sentry-junior-source-1/content.md) ### Lesson Stateless compute fights you; serverless functions time out and disappear; model the agent around interrupt/resume Inference · Catalog judgment · Medium confidence · `sentry-junior--lessons-learned-2` Confidence reason: The catalog derives this observation from the linked sources. Evidence: - Supports · [1] [Building an Intern (Junior at Sentry)](https://cra.mr/building-an-intern/) · [Preserved copy](https://github.com/steel-experiments/internal-agents-map/blob/main/archive/sources/sentry-junior-source-1/content.md) ### Lesson Unit tests are the wrong yardstick for agents; invest in evals and integration tests instead Inference · Catalog judgment · Medium confidence · `sentry-junior--lessons-learned-3` Confidence reason: The catalog derives this observation from the linked sources. Evidence: - Supports · [1] [Building an Intern (Junior at Sentry)](https://cra.mr/building-an-intern/) · [Preserved copy](https://github.com/steel-experiments/internal-agents-map/blob/main/archive/sources/sentry-junior-source-1/content.md) ### Lesson Writes need per-user authorization, not blanket trust Inference · Catalog judgment · Medium confidence · `sentry-junior--lessons-learned-4` Confidence reason: The catalog derives this observation from the linked sources. Evidence: - Supports · [1] [Building an Intern (Junior at Sentry)](https://cra.mr/building-an-intern/) · [Preserved copy](https://github.com/steel-experiments/internal-agents-map/blob/main/archive/sources/sentry-junior-source-1/content.md) ## Sources 1. [Building an Intern (Junior at Sentry)](https://cra.mr/building-an-intern/) - Engineering blog · First party · Evidence - Original URL: - Accessed: 2026-08-12 · Last verified: 2026-08-31 - Preserved copy in the repository: 2. [getsentry/junior source repository](https://github.com/getsentry/junior) - Repository · First party · Evidence - Original URL: - Accessed: 2026-08-12 · Last verified: 2026-08-31 - Preserved copy in the repository: 3. [Sentry Labs](https://labs.sentry.dev/) - Documentation · First party · Evidence - Original URL: - Accessed: 2026-08-12 · Last verified: 2026-08-31 - Preserved copy in the repository: 4. [getsentry/junior Apache 2.0 license](https://github.com/getsentry/junior/blob/main/LICENSE?plain=1) - Source code · First party · Evidence - Original URL: - Accessed: 2026-08-31 · Last verified: 2026-08-31 - Preserved copy in the repository: