CtrlK
BlogDocsLog inGet started
Tessl Logo
SkillPublic

quintet-discipline

Verification and intent hygiene for multi-agent team and fleet execution. Use when decomposing tasks across quintet worker teams or verifying work produced by background providers (where workers self-report 'DONE' or outputs may be hallucinated/empty). Focuses on verifying real file artifacts and diffs instead of blindly trusting self-reports. Does NOT mandate running test suites for routine code edits, config tweaks, or version bumps.

Quality

77%

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/quintet-discipline/SKILL.md
SKILL.md
Quality
Evals
Security

Quality

Content

65%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 concise, actionable skill body with a useful proportionality table and explicit pre-shutdown verification checkpoint. Its two real gaps are the missing recovery path when artifact verification fails, and complete silence about the persona reference files shipped in the bundle.

Suggestions

Add a short feedback-loop step for failed verification (e.g., 'if a worker's output file is empty or the diff shows no real change: re-prompt that worker or reassign the task before proceeding') to raise workflow clarity for this batch-operation skill.

Add a References section linking references/personas/debugger.md, devops-troubleshooter.md, and security-auditor.md (e.g., 'For specialized worker prompts, see references/personas/') so the bundled files are discoverable.

Consider consolidating the verification checks into one copy-paste-ready command sequence per scope (e.g., `wc -c <files> && git diff --stat`) to make the guidance fully executable.

DimensionReasoningScore

Conciseness

Mostly lean, table-driven guidance at ~47 lines; only minor over-explanation of things Claude already knows ('Two workers writing to the same file will overwrite or clobber each other', 'pane histories and unsaved state are gone') that could be trimmed.

4 / 5

Actionability

Concrete commands appear throughout (`git diff`, `git status`, `wc -c`, `quintet team shutdown <name>`) with clear per-scope checks, but they are embedded in table cells and prose rather than as ready-to-run verification sequences.

4 / 5

Workflow Clarity

The dispatch section has a clear 3-step sequence with an explicit checkpoint ('Inspect before shutdown'), but this is a batch/multi-worker skill with no validate→fix→retry feedback loop — what to do when verification fails (empty file, hallucinated diff) is never addressed, capping this dimension per the rubric.

3 / 5

Progressive Disclosure

The body is well-organized into three clear sections, but the bundle's references/personas/ files (debugger.md, devops-troubleshooter.md, security-auditor.md) are never mentioned anywhere in SKILL.md, leaving provided reference material undiscoverable.

3 / 5

Total

14

/

20

Passed

Description

78%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 description with an explicit 'Use when' clause, concrete capability statements, third-person voice, and a well-drawn negative boundary ('Does NOT mandate running test suites'). The main weakness is the vague 'intent hygiene' phrase, which names a core purpose without explaining it.

DimensionReasoningScore

Specificity

Lists several explicit actions ('decomposing tasks across quintet worker teams', 'verifying real file artifacts and diffs', 'Does NOT mandate running test suites'), though the opening 'Verification and intent hygiene' is abstract and the intent/scope guidance from the body is not surfaced.

4 / 5

Completeness

Both what ('verifying real file artifacts and diffs instead of blindly trusting self-reports') and when ('Use when decomposing tasks across quintet worker teams or verifying work produced by background providers') are explicit; held below 5 because 'intent hygiene' — half of the stated what — is never explained.

4 / 5

Trigger Term Quality

Good natural keyword coverage ('multi-agent team', 'fleet execution', 'background providers', "self-report 'DONE'", 'hallucinated/empty'), but common synonyms such as 'subagents' or 'parallel agents' are missing.

4 / 5

Distinctiveness Conflict Risk

Niche triggers ('quintet worker teams', 'background providers', 'fleet execution', "workers self-report 'DONE'") clearly distinguish this from general code-review or testing skills with minimal conflict risk.

5 / 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

salemaziel/omc-octo-quintet

Path

skills/quintet-discipline/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.