OpenAgentSkill Resolve API
Task: find the right skill
Policy: human_review_required

Recommendation:
Best skill: Playwright Browser Skill (openai-playwright)
URL: https://www.openagentskill.com/skills/openai-playwright
Repository: https://github.com/openai/skills/tree/main/skills/.curated/playwright
Install: npx skills add openai/skills --skill playwright
Install policy: review
Trust Score v5: 83/100 Review then install
Trust decision: Ask for approval or run a sandbox-only trial before installing.
Risk: Safe to try; Reviewed with permission notes; 86/100 Production candidate
Why:
- Matches task terms: the, right, skill
- Strong GitHub adoption: 24,000 stars
- Quality score 91/100
- Install handoff is available
- Repository freshness signal is available
- Primary pick for Browser automation
- 83/100 OpenAgentSkill Trust Score v5
- 91/100 audit score
Agent instruction:
Use Playwright Browser Skill for: find the right skill
Review audit: https://www.openagentskill.com/skills/openai-playwright/audit
Ask for human approval before installing: npx skills add openai/skills --skill playwright
If the audit or policy warnings are unacceptable, choose one of the alternatives.
Agent contract:
- Version: openagentskill-resolve-contract-v2
- Install policy: review
- Auto-install allowed: no
- Human review required: yes
- Audit URL: https://www.openagentskill.com/skills/openai-playwright/audit
- Do not use when:
  - teams that need a vendor-supported SLA
  - production agents without a repository review
  - Browser automation can submit data; confirm the target environment and review actions before execution.
  - No OpenAgentSkill engagement data yet

Feedback:
Event ID: resolve_263d4339-93e7-41a6-a500-4b2453a79be1
Outcome API: https://www.openagentskill.com/api/agent/outcome
CLI: npx openagentskill outcome "resolve_263d4339-93e7-41a6-a500-4b2453a79be1" --skill "openai-playwright" --task "find the right skill" --agent "auto" --outcome success

Agent Feedback Loop:
Version: openagentskill-agent-feedback-loop-v3
Dry run: supported
Expected outcomes: success, failed, not_relevant, blocked_by_risk, setup_required
Error types: install_failed, runtime_error, not_relevant, permission_blocked, risk_blocked, setup_required, low_quality_output, missing_dependency, timeout, other
Workspaces: sandbox, local, ci, production, unknown
Updates: Trust Score v5 outcome confidence; Resolve ranking outcome evidence; Skill detail outcome signal block; Outcome leaderboard

Install Receipt:
OpenAgentSkill Install Receipt
Version: openagentskill-install-receipt-v1
Receipt ID: receipt_263d4339-93e7-41a6-a500-4b2453a79be1
Resolve event: resolve_263d4339-93e7-41a6-a500-4b2453a79be1
Generated: 2026-08-03T00:41:09.034Z

Task: find the right skill
Agent: auto

Selected skill: Playwright Browser Skill (openai-playwright)
URL: https://www.openagentskill.com/skills/openai-playwright
Repository: https://github.com/openai/skills/tree/main/skills/.curated/playwright
Audit: https://www.openagentskill.com/skills/openai-playwright/audit
Eval: https://www.openagentskill.com/api/agent/evals?slug=openai-playwright&task=find%20the%20right%20skill&max_risk=medium

Install: npx skills add openai/skills --skill playwright
Install policy: review
Auto-install allowed: no
Human review required: yes
Sandbox first: yes

Trust: 83/100 Review then install
Agent Proven: 0/100 Needs first agent run
Audit: 91/100 Safe to try
Safety: 63/100 Reviewed with permission notes
Outcome signal: No agent outcome data yet

Why selected:
- Matches task terms: the, right, skill
- Strong GitHub adoption: 24,000 stars
- Quality score 91/100
- Install handoff is available
- Repository freshness signal is available
- Primary pick for Browser automation
- 83/100 Trust Score
- 0/100 Agent Proven Score

Do not use when:
- teams that need a vendor-supported SLA
- production agents without a repository review
- Browser automation can submit data; confirm the target environment and review actions before execution.
- No OpenAgentSkill engagement data yet
- High-risk permission hints: Shell or command execution
- Production credentials, payments, or irreversible account changes without explicit human review
- Sensitive private data before reviewing repository code, license, and permission surface

Alternatives:
- Webapp Testing (anthropic-webapp-testing): npx skills add anthropics/skills --skill webapp-testing
- RNSkill Video Title (pluviobyte-ra-video-title): npx -y skills add Pluviobyte/rnskill --skill ra-video-title
- RNSkill Practical Video Planning (pluviobyte-ra-practical-video-planning): npx -y skills add Pluviobyte/rnskill --skill ra-实操策划
- RNSkill Editorial Video Cover (pluviobyte-rn-cover-skill): npx -y skills add Pluviobyte/rnskill --skill rn-cover-skill

Outcome feedback:
Endpoint: POST https://www.openagentskill.com/api/agent/outcome
Event ID: resolve_263d4339-93e7-41a6-a500-4b2453a79be1
CLI: npx openagentskill outcome "resolve_263d4339-93e7-41a6-a500-4b2453a79be1" --skill "openai-playwright" --task "find the right skill" --agent "auto" --outcome success

Next steps:
1. Read the audit page: https://www.openagentskill.com/skills/openai-playwright/audit
2. Fetch the pre-install eval: https://www.openagentskill.com/api/agent/evals?slug=openai-playwright&task=find%20the%20right%20skill&max_risk=medium
3. Do not auto-install yet. Ask for human review before running: npx skills add openai/skills --skill playwright
4. Run one narrow task with no production secrets or irreversible side effects.
5. Report the result to https://www.openagentskill.com/api/agent/outcome using event_id resolve_263d4339-93e7-41a6-a500-4b2453a79be1.

Decision Packet:
Version: openagentskill-agent-decision-packet-v1
Selected: Playwright Browser Skill (openai-playwright)
Install: npx skills add openai/skills --skill playwright
Policy: review
Trust: 83/100 Review then install
Trust decision: Ask for approval or run a sandbox-only trial before installing.
Risk: Safe to try; Reviewed with permission notes
Outcome endpoint: https://www.openagentskill.com/api/agent/outcome
Outcome event: resolve_263d4339-93e7-41a6-a500-4b2453a79be1
Do not use when:
- teams that need a vendor-supported SLA
- production agents without a repository review
- Browser automation can submit data; confirm the target environment and review actions before execution.
- No OpenAgentSkill engagement data yet
Alternatives:
- Webapp Testing: npx skills add anthropics/skills --skill webapp-testing
- RNSkill Video Title: npx -y skills add Pluviobyte/rnskill --skill ra-video-title
- RNSkill Practical Video Planning: npx -y skills add Pluviobyte/rnskill --skill ra-实操策划

Selected:
Playwright Browser Skill (openai-playwright)
Match: 99
Supply: Coding / Testing and QA
Maintenance: 19d since push
Risk: Safe to try
Safety: 63/100 Review before install
Safety tier: Reviewed with permission notes
Auto-install policy: review
Audit: 91/100 Safe to try
Install: npx skills add openai/skills --skill playwright
URL: https://www.openagentskill.com/skills/openai-playwright
Why: Matches task terms: the, right, skill; Strong GitHub adoption: 24,000 stars; Quality score 91/100; Install handoff is available; Repository freshness signal is available

Agent Decision:
Recommended skill: Playwright Browser Skill
Install command: npx skills add openai/skills --skill playwright
Safety gate: Reviewed with permission notes (review)
Why recommended:
- Matches task terms: the, right, skill
- Strong GitHub adoption: 24,000 stars
- Quality score 91/100
- Install handoff is available
- Repository freshness signal is available
- Primary pick for Browser automation
Risk summary: Safe to try; Reviewed with permission notes; 63/100 Review before install; 86/100 Production candidate
Risk notes:
- High-risk permission hints: Shell or command execution
- Browser automation can submit data; confirm the target environment and review actions before execution.
Next steps:
- Read the audit URL before installing.
- Ask for human approval before installing.
- Use install command: npx skills add openai/skills --skill playwright
- After one narrow task, report the outcome with: npx openagentskill outcome "resolve_263d4339-93e7-41a6-a500-4b2453a79be1" --skill "openai-playwright" --task "find the right skill" --agent "auto" --outcome success
- Summarize output, warnings, files touched, and whether setup or human review was required.

Agent Workflow:
Action: Require human approval before installing into a real workspace.
Install API: https://www.openagentskill.com/api/skills/openai-playwright/install
Command: npx skills add openai/skills --skill playwright
Prompt:
Task: find the right skill
Use Playwright Browser Skill from https://www.openagentskill.com/skills/openai-playwright.
Review the audit first: https://www.openagentskill.com/skills/openai-playwright/audit
Review the eval first: https://www.openagentskill.com/api/agent/evals?slug=openai-playwright&task=find%20the%20right%20skill&max_risk=medium
Install handoff: https://www.openagentskill.com/api/skills/openai-playwright/install
Install command: npx skills add openai/skills --skill playwright
If audit or policy warnings look unsafe for this workspace, use one of the alternatives instead.

Review checklist:
- Safety tier: Reviewed with permission notes
- Safety score: 63/100 Review before install
- Audit score: 91/100 Safe to try
- Trust score: 83/100 Review then install
- Readiness: 100/100 Production-ready
- High-risk permission hints: Shell or command execution
- Browser automation can submit data; confirm the target environment and review actions before execution.
- Browser automation can submit data; confirm the target environment and review actions before execution.

Agent Handoff:
Version: openagentskill-agent-handoff-v1
Mode: resolve_compare_review_install
Selected skill: Playwright Browser Skill (openai-playwright)
Install policy: review
Auto-install allowed: no
Human review required: yes
Platform templates:
- Codex: use copy_prompt from agent_handoff.platform_templates.codex
- Claude Code: use copy_prompt from agent_handoff.platform_templates.claude-code
- Cursor: use copy_prompt from agent_handoff.platform_templates.cursor
Blocked actions:
- Do not install when safety_gate.blocked is true.
- Do not install when the audit or eval reports unacceptable workspace risk.
- Do not execute shell commands, access secrets, or call external services without user approval.
- Do not keep the skill as an always-on instruction if it only fits a narrow task.

Alternatives:
2. Webapp Testing (anthropic-webapp-testing) — Reviewed, safety 77/100, audit 93/100
3. RNSkill Video Title (pluviobyte-ra-video-title) — Reviewed with permission notes, safety 76/100, audit 88/100
4. RNSkill Practical Video Planning (pluviobyte-ra-practical-video-planning) — Reviewed with permission notes, safety 72/100, audit 88/100
5. RNSkill Editorial Video Cover (pluviobyte-rn-cover-skill) — Reviewed with permission notes, safety 76/100, audit 88/100

Blocked for auto-install:
No blocked candidates