Tessl • The management layer for AI agent skills

SkillPublic

do-in-steps

Execute one complex task as ordered, dependent steps run sequentially, passing context from each step to the next, with per-step LLM-as-a-judge verification. Use when later steps depend on the results of earlier ones.

Quality

75%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

Securityby

Passed

No findings from the security scan

Fix and improve this skill with Tessl

tessl review fix ./plugins/sadd/skills/do-in-steps/SKILL.md

Quality

Content

66%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.

The skill is highly actionable with a flawlessly sequenced, validation-gated workflow. Its main weaknesses are severe verbosity from rule restatement and duplicated example prompts, and a monolithic structure that should be split across reference files.

Suggestions

Split the three worked Examples, the prompt templates, and the Model Selection Policy into separate reference files linked from SKILL.md.

De-duplicate the parallel-dispatch and meta-judge-reuse rules — state each once authoritatively and reference it elsewhere.

Trim the full reprinted judge prompts in the Examples to the parts that differ from the section 3.4 template.

Dimension	Reasoning	Score
Conciseness	At ~1500 lines the body is noticeably verbose: the parallel-dispatch, "never re-run meta-judge", and "never disclose the threshold" rules are each restated 4-6 times, and the three full Examples reprint entire judge prompts that duplicate the section 3.4 template.	2 / 5
Actionability	Provides fully executable, copy-paste-ready material: CoT/self-critique/judge/retry prompt templates, model-selection and decomposition tables, the `mkdir -p .specs/reports` command, report naming convention, and concrete Task-tool dispatch examples.	5 / 5
Workflow Clarity	Phases 1-4 are explicitly sequenced with validation checkpoints (judge verify, score bands ≥4.0 / 3.0-4.0 / <3.0), a max-3 retry feedback loop, and an escalation ladder with error-recovery guidance.	5 / 5
Progressive Disclosure	No bundle files exist and the entire 1500-line skill is inlined in SKILL.md; the giant examples, prompt templates, and model-selection policy clearly belong in separate reference files but are not split out.	2 / 5
	Total	14 / 20 Passed

Description

76%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 clear and complete, answering both what and when with a concrete trigger and a distinct niche. It is slightly held back by trigger phrasing that is more technical than what a user would naturally say and by minor gaps in the action inventory.

Suggestions

Add natural trigger phrases users actually say, e.g. "Use for multi-step or sequential tasks where later steps depend on earlier ones".

Mention decomposition and per-step model selection so the concrete action list is more comprehensive.

Dimension	Reasoning	Score
Specificity	Lists several concrete actions — "ordered, dependent steps run sequentially", "passing context from each step to the next", "per-step LLM-as-a-judge verification" — but omits decomposition and model selection, leaving minor coverage gaps.	4 / 5
Completeness	Clearly states what (sequential dependent steps with context passing and per-step judge verification) and an explicit when clause with a concrete trigger ("Use when later steps depend on the results of earlier ones").	5 / 5
Trigger Term Quality	The trigger "Use when later steps depend on the results of earlier ones" is relevant but technical; common natural phrasings like "step by step", "sequential", or "multi-step task" are missing.	3 / 5
Distinctiveness Conflict Risk	The sequential-dependent-steps-with-judge-verification niche is mostly distinct, with only minor overlap risk against other generic multi-step orchestration skills.	4 / 5
	Total	16 / 20 Passed

Validation

87%

Warnings & errors only

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

Validation — 14 / 16 Passed

Validation for skill structure

Criteria	Description	Result
skill_md_line_count	SKILL.md is long (1503 lines); consider splitting into references/ and linking	Warning
frontmatter_unknown_keys	Unknown frontmatter key(s) found; consider removing or moving to metadata	Warning

	Total	14 / 16 Passed

Repository: NeoLabHQ/context-engineering-kit
Path: plugins/sadd/skills/do-in-steps/SKILL.md
Revision: 23e2428

Reviewed: 20 days ago

Discovery Implementation Validation

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.

do-in-steps

Table of Contents

Is this your skill?