CtrlK
BlogDocsLog inGet started
Tessl Logo

subagent-driven-development

Execute plans via delegate_task subagents (2-stage review).

57

Quality

67%

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

Fix and improve this skill with Tessl

tessl review fix ./optional-skills/software-development/subagent-driven-development/SKILL.md
SKILL.md
Quality
Evals
Security

Quality

Content

81%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 body is an exemplary operational guide: fully executable dispatch examples, a clearly gated two-stage review workflow with feedback loops, and well-signaled one-level references. Its main weakness is redundancy — the core rules are repeated across several closing sections, which inflates token cost without adding guidance.

Suggestions

Consolidate the overlapping rule statements in 'Red Flags', 'Handling Issues', 'Efficiency Notes', and 'Remember' into a single canonical checklist to cut repeated tokens.

Trim the 'Example Workflow' narrative to one task's full loop (spec fail → fix → re-review) instead of two plus filler ('... continue for all tasks').

Consider moving the 'Integration with Other Skills' section into a reference file, keeping the body focused on the dispatch/review loop itself.

DimensionReasoningScore

Conciseness

Mostly efficient with concrete examples throughout, but the same rules are restated across 'Red Flags', 'Handling Issues', 'Efficiency Notes', 'Remember', and the lengthy 'Example Workflow' (e.g. 'Never skip reviews' / 'Skip review loops' / 'Never skip reviews' appear in multiple sections). It could be tightened without losing guidance, so it sits at the 'mostly efficient but could be tightened' anchor rather than 4.

3 / 5

Actionability

Fully concrete, copy-paste-ready guidance: complete delegate_task(...) calls with goal, full context strings, and toolsets; todo(...) invocations; runnable commands ('pytest tests/ -q', 'git add -A && git commit -m ...') and a worked example covering the common case. Matches the 'fully executable' anchor.

5 / 5

Workflow Clarity

The process is clearly sequenced (parse plan → dispatch implementer → spec review → quality review → mark complete → final integration review → verify/commit) with explicit validation gates ('Continue only when spec-compliant', 'Continue only when approved'), feedback loops ('Repeat until approved', 'Don't skip the re-review'), ordering rules, and checklists. Matches the top anchor including feedback loops for error recovery.

5 / 5

Progressive Disclosure

Good structure with well-signaled, one-level-deep references — both 'references/context-budget-discipline.md' and 'references/gates-taxonomy.md' exist and each has an explicit load condition ('Load when a run will clearly consume significant context', 'Load when designing or reviewing any workflow that has validation checkpoints'). It falls short of the top anchor because the ~330-line body inlines sizable narrative content (Example Workflow, Integration with Other Skills) that pushes it beyond a lean overview.

4 / 5

Total

17

/

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 concise and names a distinct capability, but it omits any 'when to use' trigger guidance and lacks natural trigger-term coverage. It reads more like a terse title than a discovery-ready description.

Suggestions

Add an explicit 'Use when...' clause, e.g. 'Use when you have an implementation plan with multiple tasks to execute, or when the user asks to implement a plan with automated review between tasks.'

Include natural trigger terms users would say: 'implementation plan', 'execute tasks', 'delegate', 'parallel implementation', 'code review between tasks'.

Enumerate the concrete actions (implement each task via subagent, spec-compliance review, code-quality review, final integration review) to raise specificity.

DimensionReasoningScore

Specificity

Names the domain and 1-2 concrete actions ('Execute plans via delegate_task subagents', '2-stage review') but coverage is not comprehensive — no mention of implementing tasks, review loops, or fixing cycles. It does not list several specific actions, so it falls below anchor 4.

3 / 5

Completeness

Has a clear 'what' (execute plans via subagents with two-stage review) but no 'Use when...' clause or equivalent trigger guidance, which caps completeness at 3 per the judging guidelines.

3 / 5

Trigger Term Quality

Relevant keywords like 'plans', 'subagents', and 'delegate_task' are present, but common natural variations users would say ('implementation plan', 'execute tasks', 'delegation', 'parallel work') are missing, matching the 'some relevant keywords but missing variations' anchor.

3 / 5

Distinctiveness Conflict Risk

The 'delegate_task' and '2-stage review' framing carves a fairly distinct niche with only minor overlap risk against closely related skills like 'plan' and 'requesting-code-review'; it is not fully conflict-free since plan execution/review boundaries are adjacent.

4 / 5

Total

13

/

20

Passed

Validation

81%

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

Validation — 13 / 16 Passed

Validation for skill structure

CriteriaDescriptionResult

metadata_version

'metadata.version' is missing

Warning

metadata_field

'metadata' should map string keys to string values

Warning

frontmatter_unknown_keys

Unknown frontmatter key(s) found; consider removing or moving to metadata

Warning

Total

13

/

16

Passed

Repository
NousResearch/hermes-agent
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.