CtrlK
BlogDocsLog inGet started
Tessl Logo
SkillPublic

flow-parallel

Decompose and execute large changes, migrations, or multi-issue fixes in parallel with quality gates

Quality

70%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide
SecuritybySnyk

Low

Low-risk findings worth noting

Fix and improve this skill with Tessl

tessl review fix ./skills/flow-parallel/SKILL.md
SKILL.md
Quality
Evals
Security

Quality

Content

62%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.

An exceptionally actionable, well-sequenced orchestration workflow with real validation gates and error-recovery loops, undermined by monolithic inline scripts and heavy repetitive emphasis that inflate token cost. Moving the launch script and templates into bundle files and defining the initial BASE_REF selection would resolve the main weaknesses.

Suggestions

Move the ~120-line launch.sh template (and the banner/host-adapter text) into a scripts/ or references/ file and reference it one level deep, cutting the body to a navigable overview.

Specify how to select BASE_REF in Step 5 (e.g. 'use the current trunk branch: BASE_REF=$(git rev-parse --abbrev-ref HEAD) or the commit HEAD of the shared starting point') so the script is fully executable.

Trim redundancy: the Prohibitions list and repeated 'DO NOT PROCEED' gates largely restate the per-step MANDATORY blocks — consolidate into one checklist.

DimensionReasoningScore

Conciseness

The ~900-line body is noticeably verbose: the 'Prohibitions' section restates the step rules already enforced inline, 'DO NOT PROCEED TO STEP N' gates and 'MANDATORY/CANNOT' emphasis repeat throughout, and the fixed banner template is padded. It avoids score 1 because it does not explain concepts Claude already knows — the bulk is operational script content rather than padding.

2 / 5

Actionability

Mostly executable end-to-end: complete bash launch scripts, a runnable Python dependency/cycle validator, per-step validation gates, and explicit placeholder-replacement instructions. It falls short of fully copy-paste ready because Step 5 opens with ': "${BASE_REF:?Select the shared base ref...}"' without ever defining how to choose BASE_REF, leaving a key input to guesswork.

4 / 5

Workflow Clarity

The 7-step sequence is explicit with named validation gates after each step (wbs_generated, instructions_written, processes_launched, all_work_packages_complete), blocking rules between steps, and a dedicated error-handling section with retry/timeout feedback loops for this batch operation. This matches the anchor for clear sequencing with explicit validation and error-recovery loops.

5 / 5

Progressive Disclosure

In-file structure is good (numbered steps, clear section headers), but the ~120-line launch.sh template, the banner text, and the host-adapter notes clearly belong in separate bundle files, and no references/, scripts/, or assets/ exist — the only file reference (skills/blocks/codex-host-adapter.md) points outside the bundle. This sits between the 'content that should be separate is inline' anchor and the 'minimal structure' anchor.

3 / 5

Total

14

/

20

Passed

Description

66%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 concrete, domain-specific description that clearly states what the skill does, weakened by the complete absence of any 'when to use this' trigger guidance. Adding an explicit 'Use when...' clause with natural trigger phrases would lift both completeness and distinctiveness.

Suggestions

Append an explicit trigger clause, e.g. 'Use when the user asks to parallelize a large refactor, run a migration in batches, fix multiple issues at once, or split a big task into independent work packages.'

Include common user phrasings like 'parallelize', 'batch', or 'break this down into pieces' as trigger terms alongside the current vocabulary.

Sharpen 'quality gates' into a concrete signal (e.g. 'with per-package validation and exit-code checks') to reduce overlap with generic orchestration skills.

DimensionReasoningScore

Specificity

The description names the domain ("large changes, migrations, or multi-issue fixes") and several concrete actions ("Decompose and execute ... in parallel with quality gates"), matching the anchor that lists several specific actions with minor coverage gaps. It falls short of 5 because 'quality gates' and the parallel-execution mechanics are not spelled out, and above 3 because more than 1-2 concrete actions are given.

4 / 5

Completeness

The 'what' is clear (decompose and execute compound tasks in parallel with quality gates), but there is no 'Use when...' clause or equivalent explicit trigger guidance — the cap for a missing when-clause applies. It is not 4 because the 'when' is entirely absent rather than merely implicit or imprecise, and not 2 because the 'what' half is concrete and specific.

3 / 5

Trigger Term Quality

Natural phrases like "migrations", "multi-issue fixes", and "in parallel" match what users would actually say, giving good keyword coverage. Common variations such as 'parallelize', 'batch', 'break down this refactor', or 'work packages' are missing, keeping it below the comprehensive-synonyms anchor.

4 / 5

Distinctiveness Conflict Risk

The decomposition-plus-parallel-execution-plus-quality-gates combination carves a fairly distinct orchestration niche with minor overlap risk against general coding or code-review skills. It is below 5 because 'execute large changes' is broad enough to overlap with plain implementation skills lacking a more distinctive trigger.

4 / 5

Total

15

/

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

skill_md_line_count

SKILL.md is long (926 lines); consider splitting into references/ and linking

Warning

Total

15

/

16

Passed

Repository

nyldn/claude-octopus

Path

skills/flow-parallel/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.