Repository navigation
managed-agents: map the docs corpus and pick a first credit-funded workload #7675
Copy link
Copy link
Open
Labels
priority: mediumReal value, no hard deadline; normal backlog flow.Real value, no hard deadline; normal backlog flow.status: needs-decisionAwaiting a human or maintainer judgment call.Awaiting a human or maintainer judgment call.work-class: scopedA briefed fix or small feature; blast radius bounded by the brief, tests exist.A briefed fix or small feature; blast radius bounded by the brief, tests exist.
Description
Refs: #6958
Context
The claude.dev post "Building effective agent automations" (2026-10-08, https://claude.dev/blog/building-effective-agent-automations/) runs a scheduled agent on Managed Agents and points readers at an onboarding path. Anthropic's credit pages say the monthly API credits cover Managed Agents. Managed Agents is in beta and is not eligible for ZDR or a HIPAA BAA. Digest entries 18 and 19.
Decision
- Q11: entry 18 moves forward as its own track instead of deferring.
- Q12: digest the Managed Agents docs with a
/knowledge:map-corpusrun scoped to the platform.claude.com managed-agents pages, then queue its pages for/knowledge:docpage-digest. - Q38: credit-funded work is backlog work with value of its own; this track is the first Managed Agents candidate. Work is never created only to spend credit.
- Entry 19: the operator decides whether to run the onboarding path; fetch the post's reference
agent.mdonly as part of this track.
Acceptance criteria
- A verified corpus map of the Managed Agents docs exists, with an approved digest queue.
- One first workload is chosen, with its basis: it needs no local machine, it is read-only or its writes are scoped, and its data is fit for a beta surface that is not ZDR or HIPAA BAA eligible.
- The choice states how a run's failure is detected, given that a succeeded deployment run only means a session was created.
- No price, rate or per-task cost appears in tracked docs outside the cost-claims exceptions.
Related
- Managed Agents: workload fit, credit coverage of runtime, and how a session gets repo access, skills, MCP servers and network #6969 (closed): settled whether Managed Agents fits our workloads; this issue acts on that fit.
- Allocation model: which use case runs on which surface and pool, with a filler that drains whichever budget expires next #6970: allocation model across surfaces and pools; this issue supplies its first Managed Agents workload.
Out of scope
- Moving any existing lane onto Managed Agents.
Reactions are currently unavailable
Metadata
Metadata
Assignees
Labels
priority: mediumReal value, no hard deadline; normal backlog flow.Real value, no hard deadline; normal backlog flow.status: needs-decisionAwaiting a human or maintainer judgment call.Awaiting a human or maintainer judgment call.work-class: scopedA briefed fix or small feature; blast radius bounded by the brief, tests exist.A briefed fix or small feature; blast radius bounded by the brief, tests exist.
Type
Fields
No fields configured for Task.
Activity
github-actions commentedon Oct 16, 2026
Proposed work class: work-class: scoped
Basis: automated intake triage; the lanes App applied this class.
Size: Medium
Complexity: Medium
Uncertainty: Knowable
Flags: none
Blocked by: none
Understood: run a corpus map of the Managed Agents docs, queue the pages for digest, and pick one first workload. The workload must need no local machine, be read-only or write only in a narrow scope, and use data suited to a beta surface without ZDR or HIPAA BAA eligibility. The choice must also say how a failed run is detected, since a succeeded run only means a session was created. The issue states its decisions (Q11, Q12, Q38, entry 19), so nothing is missing. I found no duplicates. Credit spend is prepaid, so there is no billing flag.
Automated triage, applied by the lanes App.
kyle-sexton commentedon Oct 25, 2026
Released for re-triage under the owner's 2026-10-24 rule: an issue stops for a person only for an intent or vision question, a security call, a host or attended run, or spend outside prepaid credit. The hold label came off and intake triage runs again now; it re-applies a hold, with a numbered question, when one of those applies.
To hold this issue again, add
needs-human(orstatus: needs-decision) and say why in a comment.melodic-automation-lanes commentedon Oct 25, 2026
Not admitted to the issue lane: no trusted-human acceptance criteria: a listed human must put the confirmed
- [ ]criteria in the body or post them as their own comment, or, on an issue a listed human filed, intake's newest refinement comment must hold them unedited. Admission takes the issue once each of those is met, when intake triages it again (.github/actions/resolve-config/admission-rules.json).melodic-automation-lanes commentedon Oct 25, 2026
The latest triage verdict holds nothing: admission moved the issue to
status: needs-decision; its comment says why.melodic-automation-lanes commentedon Oct 25, 2026
Refinement. A read-only run explored this repository, searched its issues and pull requests, and read the allowed docs to draft what follows. On an issue a listed human filed, it counts as the issue's spec once posted, while unedited; it is still model text, so check it.
Intended outcome: A verified map of the platform.claude.com Managed Agents docs with a digest queue, plus one chosen read-only first workload with a failure-detection rule, recorded in a docs file.
Dependencies
Related work
Open questions
RECOMMENDED: Treat it as approved when the map-corpus verification gate passes and the queue file is merged; the owner can strike pages in review. Basis: judgment
RECOMMENDED: Public repository data only: no secrets, no customer data, no private-repo contents beyond what lanes already read. Basis: judgment
RECOMMENDED: Write-up only: the issue says choose, and launching needs the operator's onboarding step (entry 19). Basis: judgment
RECOMMENDED: Put them under docs/upstream/ next to claude-dev-building-effective-agent-automations.md, with a task list of map verified, queue listed, workload and failure detection stated, and no prices. Basis: docs/upstream/claude-dev-building-effective-agent-automations.md
Sources read
Automated refinement by the lanes App. Answer with a comment whose first line is
**Decision recorded.**.