Stats
Actions
Tags
From claude-code-config
Provides reference sheets for designing safe, observable agent harnesses: tool risk taxonomy, permission decisions, budgets, evals, trust labels, event models, and streaming patterns. Load when building agent SDK apps, custom orchestrators, MCP servers, or tool-calling systems.
How this skill is triggered — by the user, by Claude, or both
Slash command
/claude-code-config:agent-harness-designThe summary Claude sees in its skill listing — used to decide when to auto-load this skill
Ten operational reference sheets for designing a safe, observable agent harness. They are **situational** — load only the one(s) relevant to the current task from `references/` (this is why they live in a skill rather than always-on rules: building an agent harness is occasional, so the detail should not bloat every session's context).
references/agent-approval-records.mdreferences/agent-budgets.mdreferences/agent-evals.mdreferences/agent-event-model.mdreferences/agent-observability.mdreferences/agent-plan-artifact.mdreferences/agent-skill-install-checklist.mdreferences/agent-streaming.mdreferences/agent-tool-design.mdreferences/context-trust-labels.mdTen operational reference sheets for designing a safe, observable agent harness. They are situational — load only the one(s) relevant to the current task from references/ (this is why they live in a skill rather than always-on rules: building an agent harness is occasional, so the detail should not bloat every session's context).
references/agent-tool-design.md — 15-class risk taxonomy, 7-type permission decision object, draft/commit naming, structured tool results, deferred tool loading, hosted vs client tools, connector code-execution pattern.references/context-trust-labels.md — trusted / semi_trusted / untrusted labels + verbatim boundary statement; prompt-injection defense.references/agent-budgets.md — 10 mandatory budget types every agent loop must declare.references/agent-evals.md — 13 eval categories + 13 adversarial test cases + when to add regression evals.references/agent-observability.md — 16 trace fields per model call, 7-question audit, 6-step incident response.references/agent-plan-artifact.md — planning mode, plan artifact format (10 fields), plan-validate-execute.references/agent-approval-records.md — approval request/result JSON schemas, scope/expiration, no self-approval.references/agent-streaming.md — buffering for incremental tool calls when stream=True; abort handling; output guardrail modes.references/agent-event-model.md — 13 typed events for harness state persistence (replay/audit/compaction/evals).references/agent-skill-install-checklist.md — pre/during/post install + audit + incident response for 3rd-party skills.Source: distilled from the agents-best-practices skill (Denis Sergeevitch, MIT) + Anthropic harness-design engineering. Read the specific reference before applying — do not work from this index alone.
npx claudepluginhub anastasiyaw/claude-code-configDesigns and optimizes AI agent action spaces, tool definitions, observation formatting, and error recovery for higher completion rates.
Designs and optimizes AI agent action spaces, tool definitions, observation formats, error recovery, and context for higher task completion rates.
Designs autonomous agent harnesses with research loops, evaluation scaffolds, locked/editable surfaces, durable logs, novelty gates, pruning, rollback, and human approval boundaries.