ck:plan — independently scanned and version-tracked by SaferSkills.
SaferSkills independently audited ck:plan (Agent Skill) and scored it 100/100 (green). The audit ran 55 deterministic rules across Security, Supply Chain, Maintenance, Transparency, and Community; it found 0 high-severity and 0 lower-severity findings. The full rule-by-rule trace and per-finding evidence are below. Free, methodology-open.
Findings & checks · 0 flagged
Every scanned point with the score it earned and what moved between them.
First recorded scan — no prior version to compare against.
The primary manifest — the file an agent reads to learn what this artifact does.
Create detailed technical implementation plans through research, codebase analysis, solution design, and comprehensive documentation.
IMPORTANT: Before you start, scan unfinished plans in the current project at ./plans/ directory, read the plan.md, if there are relevant plans with your upcoming plan, update them as well. If you're unsure or need more clarifications, use AskUserQuestion tool to ask the user.
During the pre-creation scan, detect and mark blocking relationships between plans:
plan.md frontmatter of each unfinished plan (status != completed/cancelled)blockedBy: [existing-plan-dir]blockedBy: [new-plan-dir], new plan blocks: [existing-plan-dir]blockedBy/blocksplan.md files' frontmatterAskUserQuestion with header "Plan Dependency", present detected overlap, ask user to confirm relationship type (blocks/blockedBy/none)Frontmatter fields (relative plan dir paths):
blockedBy: [260301-1200-auth-system] # This plan waits on these plans
blocks: [260228-0900-user-dashboard] # This plan blocks these plansStatus interaction: A plan with blockedBy entries where ANY blocker is not completed → plan status should note blocked in its overview. When all blockers complete, the blocked plan becomes unblocked automatically on next scan.
If invoked with a task description, proceed with planning workflow. If invoked WITHOUT arguments or with unclear intent, use AskUserQuestion to present available operations:
| Operation | Description |
|---|---|
(default) | Create implementation plan for a task |
archive | Write journal entry & archive plans |
red-team | Adversarial plan review |
validate | Critical questions interview |
Present as options via AskUserQuestion with header "Planning Operation", question "What would you like to do?".
Default: --auto (analyze task complexity and auto-pick mode).
| Flag | Mode | Research | Red Team | Validation | Cook Flag |
|---|---|---|---|---|---|
--auto | Auto-detect | Follows mode | Follows mode | Follows mode | Follows mode |
--fast | Fast | Skip | Skip | Skip | --auto |
--hard | Hard | 2 researchers | Yes | Optional | (none) |
--parallel | Parallel | 2 researchers | Yes | Optional | --parallel |
--two | Two approaches | 2+ researchers | After selection | After selection | (none) |
Add --no-tasks to skip task hydration in any mode.
Load: references/workflow-modes.md for auto-detection logic, per-mode workflows, context reminders.
Always honoring YAGNI, KISS, and DRY principles. Be honest, be brutal, straight to the point, and be concise.
Load: references/scope-challenge.md Skip if: --fast mode or trivial task (single file fix, <20 word description)
Load: references/research-phase.md Skip if: Fast mode or provided with researcher reports
Load: references/codebase-understanding.md Skip if: Provided with scout reports
Load: references/solution-design.md
Load: references/plan-organization.md
Load: references/output-standards.md
flowchart TD
A[Pre-Creation Check] --> B[Cross-Plan Scan]
B --> C[Scope Challenge]
C --> D[Mode Detection]
D -->|fast| E[Skip Research]
D -->|hard/parallel/two| F[Spawn Researchers]
E --> G[Codebase Analysis]
F --> G
G --> H[Write Plan via Planner]
H --> I{Red Team?}
I -->|Yes| J[Red Team Review]
I -->|No| K{Validate?}
J --> K
K -->|Yes| L[Validation Interview]
K -->|No| M[Hydrate Tasks]
L --> M
M --> N[Output Cook Command]
N --> O[Journal]This diagram is the authoritative workflow. Prose sections below provide detail for each node.
1b. Cross-Plan Scan → Scan unfinished plans, detect blockedBy/blocks relationships, update both plans 1c. Scope Challenge → Run Step 0 scope questions, select mode (see references/scope-challenge.md) Skip if: --fast mode or trivial task
workflow-modes.md)/ck:plan red-team {plan-path} (hard/parallel/two modes)/ck:plan validate {plan-path} (hard/parallel/two modes)--no-tasks to skip)/ck:journal to write a concise technical journal entry upon completionIMPORTANT: Invoke "/ck:project-organization" skill to organize the outputs.
./docs/development-rules.md filePlan files = persistent. Tasks = session-scoped. Hydration bridges the gap.
Default: Auto-hydrate tasks after plan files are written. Skip with --no-tasks. 3-Task Rule: <3 phases → skip task creation. Fallback: Task tools (TaskCreate/TaskUpdate/TaskGet/TaskList) are CLI-only — unavailable in VSCode extension. If they error, use TodoWrite for tracking. Plan files remain the source of truth; hydration is an optimization, not a requirement.
Load: references/task-management.md for hydration pattern, TaskCreate patterns, cook handoff protocol.
addBlockedBy chain (skip if Task tools unavailable)Check ## Plan Context injected by hooks:
Plan dir: from ## NamingAfter creating plan: node .claude/scripts/set-active-plan.cjs {plan-dir} Reports: Active plans → plan-specific path. Suggested → default path.
DO NOT create plans or reports in USER directory. MUST create plans or reports in THE CURRENT WORKING PROJECT DIRECTORY.
| Subcommand | Reference | Purpose |
|---|---|---|
/ck:plan archive | references/archive-workflow.md | Archive plans + write journal entries |
/ck:plan red-team | references/red-team-workflow.md | Adversarial plan review with hostile reviewers |
/ck:plan validate | references/validate-workflow.md | Validate plan with critical questions interview |
Remember: Plan quality determines implementation success. Be comprehensive and consider all solution aspects.
~30 seconds. Free. No account. Every finding cites a rule and a line of evidence.