CtrlK
BlogDocsLog inGet started
Tessl Logo
SkillPublic

writing-plans

Extends superpowers:writing-plans with devflow's phase-handoff at the end.

Quality

81%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide
SecuritybySnyk

Passed

No findings from the security scan

SKILL.md
Quality
Evals
Security

Quality

Content

100%Weight 40%Scale 1-5

Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.

An exemplary instruction-only extension skill: it delegates the core plan-writing process to the parent skill by name and adds only tightly scoped, fully executable additions (worktree recovery with validation/failure handling, chapter marking, terminal title, phase handoff). Structure, token efficiency, and actionability are all strong, and the inline comments capture genuinely non-obvious operational knowledge rather than padding.

DimensionReasoningScore

Conciseness

The body is lean with zero filler: every comment conveys a non-obvious gotcha Claude would not otherwise know (e.g. "spawn_task always forks off the default branch; there is no way to pin it", "Strip the leading `worktree ` token (not awk $2) so worktree paths containing spaces survive", "Never write into the throwaway spawn worktree — it is an orphan the user prunes separately"). No basic concepts are explained; it assumes Claude's competence throughout.

5 / 5

Actionability

Fully executable, copy-paste-ready guidance: a complete bash worktree-recovery block with variables, a concrete ticket regex (`[A-Z]+-[0-9]+`), an exact terminal-title escape sequence, exact invocation arguments (`--phase plan --next-phase lock-tests`), and the exact spawn title format `[<TICKET>] [MR#<N>] Lock Tests`. No pseudocode or vague direction anywhere.

5 / 5

Workflow Clarity

The sequence is explicit and ordered (Step 0 worktree re-establishment → Before actions → delegated core → After handoff), with a hard validation checkpoint and feedback loop in the recovery block: verify branch (`on=$(git branch --show-current)` ... `echo "WORKTREE RECOVERY FAILED..."; exit 1`), plus a fallback path (reuse an existing checkout when `worktree add` fails). Conditional handling (skip Step 0 when invoked manually) is also stated.

5 / 5

Progressive Disclosure

This is a short (~56-line), single-purpose extension skill with no bundle files (references/, scripts/, assets/ are absent) and no need for external references. Per the simple-skill guidance, its well-organized sections (Step 0 / Before / After) and clearly signaled one-level-deep skill references (`superpowers:writing-plans`, `devflow:phase-handoff`) fully satisfy the criterion; nothing that belongs elsewhere is inlined.

5 / 5

Total

20

/

20

Passed

Description

53%Weight 40%Scale 1-5

Based on the skill's description, can an agent find and select it at the right time? Clear, specific descriptions lead to better discovery.

The description is precise within its ecosystem but weak as a standalone trigger: it conveys the extension relationship and a clear "what" by reference, yet provides no "Use when" trigger guidance and omits the skill's own concrete actions (worktree recovery, chapter marking, lock-tests handoff). It is distinctive but would rarely be selected by a user's natural phrasing alone.

Suggestions

Add an explicit trigger clause, e.g. "Use when writing an implementation plan during a devflow phase workflow, or when a phase-handoff message asks for the plan phase."

State the skill's own concrete actions instead of relying on references: "Restores the feature worktree, writes the implementation plan, marks the Plan chapter, and hands off to the lock-tests phase."

Include natural synonyms users would say ("implementation plan", "plan phase", "devflow", "handoff") so the description matches real request phrasing.

DimensionReasoningScore

Specificity

"Extends superpowers:writing-plans with devflow's phase-handoff at the end" names the domain (writing plans, via the referenced skill) and one concrete action (appending phase-handoff), but the skill's own capabilities — worktree re-establishment, chapter marking, terminal title, the lock-tests handoff — are only available by reference and never stated. Not the minimal/generic level of "Processes PDF files" (2), but far from the comprehensive action lists of level 4-5.

3 / 5

Completeness

A "what" is stated (extends plan-writing with a terminal phase-handoff), but the "when" is entirely absent — there is no "Use when..." clause or equivalent trigger guidance, which caps completeness at 3 per the judging guidelines. It sits at the "clear what, missing when" anchor; not level 2, since the what is not vague, and not level 4, since no when is even weakly implied.

3 / 5

Trigger Term Quality

Relevant keywords are present ("writing-plans", "phase-handoff"), but common natural variations a user would actually say — "implementation plan", "write a plan", "plan phase", "devflow" — are missing, and no synonyms or alternate phrasings appear. This matches the "some relevant keywords but missing common variations or synonyms" anchor.

3 / 5

Distinctiveness Conflict Risk

Naming two exact skills ("superpowers:writing-plans", "devflow:phase-handoff") carves out a clear niche with minimal conflict risk against unrelated skills, and the "Extends X with Y" framing distinguishes it from the base writing-plans skill itself. The residual overlap risk with that closely related parent skill keeps it just below a 5.

4 / 5

Total

13

/

20

Passed

Validation

100%

Checks the skill against the spec for correct structure and formatting. All validation checks must pass before discovery and implementation can be scored.

Validation — 16 / 16 Passed

Validation for skill structure

No warnings or errors.

Repository

AndreJorgeLopes/devflow

Path

skills/writing-plans/SKILL.md

Reviewed

Table of Contents

Is this your skill?

If you maintain this skill, you can claim it as your own. Once claimed, you can manage eval scenarios, bundle related skills, attach documentation or rules, and ensure cross-agent compatibility.