plan
Shape intent into one behavior-first PlanPacket with acceptance, evidence, and write scope. Triggers: "plan", "discover and plan", "shape this goal".
How do I install this agent skill?
npx skills add https://github.com/boshu2/agentops --skill planIs this agent skill safe to install?
- Gen Agent Trust Hubpass
The 'plan' skill is a comprehensive task decomposition workflow that utilizes platform-specific tools for research and issue management. It incorporates structured planning rules and human-in-the-loop approval gates. Security considerations focus on command execution and prompt injection surfaces inherent in automated task generation.
- Socketpass
No alerts
- Snykpass
Risk: LOW · No issues
- Runlayerwarn
2/6 files flagged
- ZeroLeakspass
Score: 93/100 · 2 sections analyzed
What does this agent skill do?
Plan
Turn the caller's intent into one bounded, testable behavior. Planning owns research necessary to understand the behavior, acceptance shaping, and scope. It does not schedule, claim, assign, implement, validate, or decide readiness.
Workflow
- Restate the intent and choose one active behavior.
- Inspect only enough real context to make paths, interfaces, and evidence concrete. Existing research and specialist skills are advisory inputs.
- Write at least one normal and one edge Given/When/Then scenario.
- Name non-goals and the evidence required to judge every criterion.
- Declare
write_scope.includeandwrite_scope.exclude, including generated companions. Scope describes permitted subject content; it grants no file ownership or delivery authority. - Name
first_acceptance_checkas either an executable command or an artifact path. Commands are data for the implementer, never executed by a packet parser. - Canonically serialize
plan-packet.v1and compute its SHA-256 digest.
Optional decomposition may describe smaller behaviors, but it carries no owner, ready, claim, priority, attempt, wave, queue, lease, admission, next-action, close, release, or delivery fields.
Required output
The packet conforms to schemas/plan-packet.v1.schema.json
and contains:
- intent and acceptance digests;
- one active behavior;
- normal and edge scenarios;
- non-goals and required evidence;
- inclusive and exclusive write scope;
- one first acceptance check;
- optional advisory decomposition.
Report the packet location and digest, then stop.
How can the creator link this skill?
Add the canonical catalog link to the repository README so users can inspect current installs and available audits. The publishing guide covers the complete discovery path.
<a href="https://skillzs.dev/skills/boshu2/agentops/plan">View plan on skillZs</a>