Skill127 repo starsupdated today

agent-harness-design

Agent Harness Design provides ten modular reference sheets for building safe and observable agent systems, covering risk taxonomy, permissions, budgeting, evaluation, observability, approval workflows, streaming, event modeling, and third-party skill vetting. Load specific sheets when designing a new agent SDK application, custom orchestrator, MCP server, or implementing tool systems with structured permissions and audit requirements.

View source Repository: claude-code-config

Install in Claude Code

Copy

git clone --depth 1 https://github.com/AnastasiyaW/claude-code-config /tmp/agent-harness-design && cp -r /tmp/agent-harness-design/skills/agent-harness-design ~/.claude/skills/agent-harness-design

Then start a new Claude Code session; the skill loads automatically.

Definition

SKILL.md

# Agent Harness Design

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-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.