CtrlK
BlogDocsLog inGet started
Tessl Logo
SkillPublic

bmad-agent-architect

System architect and technical design leader. Use when the user asks to talk to Winston or requests the architect

Quality

79%

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 ./skills/bmad-agent-architect/SKILL.md
SKILL.md
Quality
Evals
Security

Quality

Content

86%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 content is a tightly written activation procedure for a persona skill: concrete commands, explicit merge semantics, error-recovery branches, and a completion checkpoint, with references kept one level deep and missing-file handling. The only meaningful gap is the absent failure path for the Step 5 config resolution, plus a few trimmable persona-flourish phrases. As an instruction-only skill its actionability is excellent.

DimensionReasoningScore

Conciseness

The body is dense operational instruction with no explanation of concepts Claude already knows, but carries minor padding: "Fully embody this persona so the user gets the best experience" and the Step 7 closing re-confirmation of steps already ordered in Steps 2 and 7. Fits anchor 4 ('minor instances of over-explanation that could be trimmed'); not 5 because of those trimmable phrases, not 3 because there is no genuinely unnecessary explanation.

4 / 5

Actionability

Guidance is fully executable: copy-paste commands ("uv run {project-root}/_bmad/scripts/resolve_customization.py --skill {skill-root} --project-root {project-root} --key agent", "npx skills add bmad-code-org/BMAD-METHOD --skill bmad"), exact three-file merge paths, and precise merge semantics ("Scalars override, tables deep-merge, arrays of tables keyed by code or id replace matching entries and append new entries"). All consumed agent keys (role, identity, communication_style, principles, persistent_facts, icon, menu, activation_steps_prepend/append) are enumerated, leaving no material gaps; fits anchor 5 rather than 4.

5 / 5

Workflow Clarity

The eight steps are clearly sequenced with explicit error recovery in Step 1 ("If the script is not found... If it fails for any other reason, resolve the agent block yourself") and a completion checkpoint ("Do not begin the main workflow until all activation steps have been completed"). Fits anchor 4 rather than 5 because Step 5's resolve_config.py invocation has no failure-handling branch, and not 3 because validation checkpoints are explicit.

4 / 5

Progressive Disclosure

A single well-organized file with clear sections (Conventions, On Activation, Steps 1–8); all references are one level deep to runtime project files ("{project-root}/_bmad/scripts/...", "{skill-root}/customize.toml") with no nested reference chains, and missing files are gracefully handled ("Any missing file is skipped"). Nothing inlined belongs in a separate file, matching anchor 5's appropriately-split, easy-navigation structure.

5 / 5

Total

18

/

20

Passed

Description

65%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 has an explicit, natural 'Use when' trigger clause and a highly distinctive persona name, but the 'what' is only a role title with no concrete capabilities. Adding specific actions and a few more trigger synonyms would lift both specificity and completeness toward the top anchors. Third-person voice is used correctly and it is appropriately concise.

Suggestions

Replace the role label with concrete actions, e.g. "Designs system architecture from ticket requirements, evaluates technology trade-offs, and produces technical design documents."

Extend the trigger clause with natural variations users would say, e.g. "...or wants help designing system architecture, choosing technologies, or reviewing technical trade-offs."

Name the key deliverables (architecture docs, tech design decisions) in the what-clause so the skill is discoverable even when the user doesn't know the Winston persona.

DimensionReasoningScore

Specificity

"System architect and technical design leader" names the domain but is a role label, not a list of concrete actions — no verbs like 'designs architecture', 'evaluates trade-offs', or 'produces design documents' appear. This matches anchor 2 ('Names the domain but actions are minimal or generic'); not 3 because no concrete actions are stated, not 1 because the domain is clearly identified.

2 / 5

Completeness

Both parts are present: the what is "System architect and technical design leader" and the when is "Use when the user asks to talk to Winston or requests the architect". The 'when' is explicit with concrete triggers, but the 'what' is a title rather than a statement of capabilities, so it fits anchor 4 ('when' could be more specific / 'what' not fully concrete); not 5 because the what lacks concrete actions, not 3 because the when is explicitly stated.

4 / 5

Trigger Term Quality

"talk to Winston" and "requests the architect" are phrasings users would naturally say, and "System architect and technical design leader" adds keyword coverage. Good but not comprehensive — natural variations like 'system design', 'architecture review', or 'help me design this system' are missing; fits anchor 4, not 5 on incomplete synonym coverage and not 3 since coverage exceeds one or two generic keywords.

4 / 5

Distinctiveness Conflict Risk

"talk to Winston" is a highly distinct persona trigger that no other skill would claim, giving a clear niche. Minor overlap risk remains: "requests the architect" could fire on generic architecture questions or sibling BMad agents, matching anchor 4 ('Mostly distinct; minor overlap risk'); not 5 due to that residual overlap, not 3 because the Winston trigger makes misfiring unlikely.

4 / 5

Total

14

/

20

Passed

Validation

93%

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

Validation — 15 / 16 Passed

Validation for skill structure

CriteriaDescriptionResult

referenced_paths_exist

Referenced path issues: 1 missing

Warning

Total

15

/

16

Passed

Repository

bmad-code-org/BMAD-METHOD

Path

skills/bmad-agent-architect/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.