CtrlK
BlogDocsLog inGet started
Tessl Logo
SkillPublic

fast-feature

Use when driving a single already-specced ticket through devflow in one session, unattended, resolving ambiguity from the codebase rather than by asking, then spec, plan, and lock-tests back-to-back with an independent reviewer standing in for the human gate, spawning a new session only for the execute phase. Also the per-ticket driver that orchestrate-epic invokes.

Quality

83%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide
SecuritybySnyk

Critical

Do not install without reviewing

SKILL.md
Quality
Evals
Security

Quality

Content

73%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.

A highly actionable, well-sequenced orchestration skill with excellent validation checkpoints and feedback loops throughout its six-step workflow. Its main cost is token redundancy: the same guardrails are restated across the intro, steps, and three separate reinforcement sections.

Suggestions

Merge the 'Common Mistakes', 'Red flags', and 'Important' sections into a single guardrails section — they currently restate the same six rules three times, costing tokens without adding new information.

State the 'ONE session, one spawn' contract once (intro or Core pattern) and let Steps 3/4/6 reference it briefly instead of re-explaining the phase-handoff suppression rationale at each step.

Provide a concrete prompt template or dispatch snippet for the Step 5 independent reviewer subagent, mirroring the executable specificity of the Step 0 bash block.

DimensionReasoningScore

Conciseness

There is no filler explaining known concepts, but core rules are repeated three to four times each — 'suppress the intermediate phase-handoff spawn' appears in the intro, the mermaid diagram, Steps 3/4/6, Common Mistakes, the Red flags table, and Important — and the Common Mistakes, Red flags, and Important sections restate the same six points, which is more than minor tightening.

3 / 5

Actionability

The body provides copy-paste-ready bash for worktree creation, exact invocations with flags ('--no-handoff', '--unattended'), and concrete reviewer-dispatch questions, with only minor gaps such as the reviewer prompt and blocker-report format being described rather than scripted.

4 / 5

Workflow Clarity

Six numbered steps plus a flowchart give a clear sequence with explicit validation checkpoints (branch confirmation in Step 0, Phase 1.5 'verify each fails for the right reason', the reviewer verdict) and a genuine feedback loop ('If it returns NEEDS_WORK, fix and re-dispatch. If it returns NEEDS_WORK twice on the same point, stop').

5 / 5

Progressive Disclosure

Sections are well organized, references to the devflow phase skills are clearly signaled, and detail is appropriately delegated to those skills rather than inlined; at ~140 lines with three overlapping reinforcement sections (Common Mistakes, Red flags, Important), some consolidation is still possible, so it fits the 'good structure, minor organization gaps' anchor.

4 / 5

Total

16

/

20

Passed

Description

83%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.

A strong, dense description that explicitly covers both when to use it and what it does, with concrete pipeline actions and clear differentiation from its sibling devflow skills. Its only weaknesses are the abstract 'driving a ticket through devflow' framing and a few missing natural trigger variations.

DimensionReasoningScore

Specificity

The description lists several concrete actions — 'resolving ambiguity from the codebase rather than by asking', 'spec, plan, and lock-tests back-to-back', 'an independent reviewer standing in for the human gate', 'spawning a new session only for the execute phase' — but 'driving a ticket through devflow' is somewhat abstract, leaving it just short of the comprehensive 5 anchor.

4 / 5

Completeness

It explicitly answers both questions: 'Use when driving a single already-specced ticket through devflow in one session, unattended' gives a concrete when, and the rest states exactly what it does (spec/plan/lock-tests in one session, reviewer-replaced gate, single execute spawn).

5 / 5

Trigger Term Quality

It contains good natural keywords for this ecosystem ('already-specced ticket', 'devflow', 'spec, plan, and lock-tests', 'unattended', 'orchestrate-epic'), though common user phrasings like 'Jira ticket', 'fast', or 'single session' variations are missing, matching the 'a few natural terms missing' anchor.

4 / 5

Distinctiveness Conflict Risk

The niche is clear (already-specced ticket, one session, unattended) and it even names its relationship to orchestrate-epic, but it sits in a family of closely related devflow phase skills it invokes (spec-feature, writing-plans, lock-tests), creating minor overlap risk.

4 / 5

Total

17

/

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/fast-feature/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.