Openclaw Agent Optimize

Use when: you want a structured audit -> options -> recommended plan to improve an OpenClaw workspace (cost, model routing, context discipline, delegation, r...

Security Scan

The skill is an advisory-only OpenClaw workspace optimization playbook whose instructions, file references, and recommended actions are consistent with its stated purpose and do not request unrelated credentials or installs.

Purpose & Capability

Name/description (agent optimization: cost, routing, context discipline) matches the contents: SKILL.md + references provide guidance for audits, cron/heartbeat changes, model-tiering, and rollback plans. The skill requests no credentials, binaries, or config paths, which is proportional for an advisory auditor.

Instruction Scope

Runtime instructions are advisory-first: perform audits, propose exact patches, and require explicit user approval before applying persistent changes (crons, agent defaults). The docs reference reading/writing workspace artifacts (memory/, references/, logs) and proposing config patches — these are expected for this purpose and are explicitly gated by an approval requirement. There are no instructions to exfiltrate secrets or to contact external endpoints.

Install Mechanism

This is instruction-only (no install spec, no code files executed). README suggests installing via a third-party ClawHub tool or copying the folder manually; those are optional distribution channels and not part of an automated install that would fetch arbitrary binaries. No archive downloads or third-party package installs are embedded in the skill itself.

Credentials

The skill declares no required environment variables, no primary credential, and no config paths. References talk about local workspace files (memory/, references/, logs) which are appropriate for an optimization/audit skill. A recommendation to consider installing an optional helper (openclaw-mem) is informational and not a demanded credential.

Persistence & Privilege

Skill metadata: always=false and user-invocable=true. SKILL.md repeatedly instructs not to mutate persistent settings or create/edit cron jobs without explicit approval and to include exact diffs, rollback, and verification steps. Autonomous invocation (model invocation allowed) is the platform default and is not combined with any unusual privileges here.

Assessment

This skill is advisory and appears internally consistent: it will read workspace artifacts (memory/, references/, logs) and produce proposed config patches and cron changes but explicitly forbids applying persistent changes without your approval. Before approving any change: review the exact diffs, confirm the rollback steps, verify file/write locations, and confirm you trust the source (ClawHub page or repository). Do not provide external secrets or grant other credentials; treat any recommendation to install additional helpers (e.g., openclaw-mem) as optional and review that project's code separately.

OpenClaw Agent Optimization

Use this skill to tune an OpenClaw workspace for cost-aware routing, parallel-first delegation, and lean context.

Default posture

This skill is advisory first. It should produce:

• audit,
• options,
• recommended plan,
• exact patch proposal,
• rollback,
• verification plan.

No persistent mutations without explicit approval.

Quick start

1. Full audit (safe, no changes)

Audit my OpenClaw setup for cost, reliability, and context bloat. Output a prioritized plan with rollback notes. Do NOT apply changes.

2. Context bloat / transcript noise

My OpenClaw context is bloating (slow replies / high cost / lots of transcript noise). Identify the top offenders (tools, crons, bootstrap files, skills) and propose the smallest reversible fixes first. Do NOT apply changes.

3. Model routing / delegation posture

Propose a model routing plan for (a) coding/engineering, (b) short notifications/reminders, (c) reasoning-heavy research/writing. Include an exact config patch + rollback plan, but do NOT apply changes.

What good output looks like

• Executive summary
• Top drivers
  • cost
  • context
  • reliability
  • operator friction
• Options A/B/C with tradeoffs
• Recommended plan (smallest safe change first)
• Exact proposals + rollback + verify

Safety contract

• Do not mutate persistent settings without explicit approval.
• Do not create/update/remove cron jobs without explicit approval.
• If an optimization reduces monitoring coverage, present options and require choice.
• Before any approved change, show:
  1. exact change,
  2. expected impact,
  3. rollback plan,
  4. post-change verification.

High-ROI optimization levers

1) Output discipline for automation

Make maintenance loops truly silent on success.

2) Separate work from notification

If you want alerts but want interactive context lean:

• do the work quietly
• notify out-of-band with a short human receipt

3) Bootstrap discipline

Keep always-injected files short and load-bearing only. Move long runbooks into references/ or adjacent notes.

4) Ambient specialist surface reduction

A common hidden tax is too many always-visible specialist skills. If a workflow is low-frequency or specialist:

• prefer on-demand worker/subagent usage,
• do not keep it permanently ambient in main-chat prompt surface.

5) Measure optimizations authoritatively

Prefer fresh-session /context json or equivalent receipts over “feels better”. High-signal fields include:

• eligible skills
• skills.promptChars
• projectContextChars
• systemPrompt.chars
• promptTokens

6) Verification-first ops hygiene

After any approved optimization, verify:

• core chat still works
• recall/behavior did not degrade
• new session actually picks up the change
• rollback path is proven, not theoretical

Workflow (concise)

1. Audit rules + memory: keep restart-critical facts only.
2. Audit skill surface: trim ambient specialists before touching tool surface.
3. Audit transcripts/noise: silence cron and heartbeat success paths.
4. Audit model routing and delegation posture.
5. Recommend the smallest viable change first.
6. Verify on a new session when skill/bootstrap snapshotting exists.

Notes

• Some runtimes snapshot skills/config per session. If you install/update skills and do not see changes, start a new session.
• Prefer short SKILL.md + references/ for long runbooks.
• If context bloat is the main complaint, pair this skill with context-clean-up (audit-only).

References

• references/optimization-playbook.md
• references/model-selection.md
• references/context-management.md
• references/agent-orchestration.md
• references/cron-optimization.md
• references/heartbeat-optimization.md
• references/memory-patterns.md
• references/continuous-learning.md
• references/safeguards.md