CtrlK
BlogDocsLog inGet started
Tessl Logo

blackbox

Delegate coding tasks to Blackbox AI CLI agent. Multi-model agent with built-in judge that runs tasks through multiple LLMs and picks the best result. Requires the blackbox CLI and a Blackbox AI API key.

56

Quality

66%

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 ./optional-skills/autonomous-ai-agents/blackbox/SKILL.md
SKILL.md
Quality
Evals
Security

Quality

Content

72%

Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.

The body is highly actionable with concrete commands and clean organization, but workflow clarity is held back by the absence of explicit validation/feedback checkpoints for long-running or parallel operations, and conciseness slips on a stray time-sensitive date. Adding verification steps and removing the date would push it toward top scores.

Suggestions

Add explicit validation/verification checkpoints for risky workflows — e.g. after a background task, verify the working tree with `git diff`/`git status` before reporting success.

Remove or generalize the hard-coded date in the checkpoint example ('task-abc123-2026-03-06') to avoid time-sensitive decay in the main body.

Tighten the redundant pty guidance (stated in both Prerequisites and Rules) into a single mention to improve token efficiency.

DimensionReasoningScore

Conciseness

The body is largely lean and command-forward, but a time-sensitive date ('task-abc123-2026-03-06') appears outside any deprecated section and the pty rationale is repeated, so it is not fully tightened to a 3.

2 / 3

Actionability

Provides fully executable terminal(...) calls, concrete flags, and copy-paste-ready command tables, matching the 'fully executable, copy-paste ready' anchor.

3 / 3

Workflow Clarity

Multi-step sequences are present (background: start -> poll/log -> submit -> kill), but there are no explicit validate-then-fix feedback loops for risky or parallel operations, which caps workflow clarity at 2.

2 / 3

Progressive Disclosure

A single well-organized SKILL.md with clearly labeled sections, tables, and no nested/deep references; for a CLI-wrapper skill the content is appropriately kept in one navigable file rather than split.

3 / 3

Total

10

/

12

Passed

Description

60%

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 concretely states what Blackbox does and its multi-model judge feature, but omits any explicit 'when to use' trigger guidance, leaving it actionable yet incomplete. Adding a 'Use when...' clause with natural trigger terms would lift both completeness and trigger-term quality.

Suggestions

Add an explicit trigger clause, e.g. 'Use when delegating a coding task to an external multi-model agent, or when you want multiple LLMs to attempt a task and pick the best result.'

Include natural user-facing keywords such as 'code generation', 'implement a feature', or 'fix a bug' alongside 'Blackbox AI' to improve trigger-term coverage.

Clarify the distinction from claude-code/codex by foregrounding the multi-model judge workflow in the description.

DimensionReasoningScore

Specificity

Lists multiple concrete actions — 'Delegate coding tasks', 'runs tasks through multiple LLMs and picks the best result' — matching the multi-action anchor rather than the single-domain anchor at 2.

3 / 3

Completeness

Clearly states what the skill does but provides no 'Use when...' clause or equivalent trigger guidance, so per the judging guidelines completeness is capped at 2 rather than 3.

2 / 3

Trigger Term Quality

Names 'Blackbox AI' and 'coding tasks' but offers no natural trigger phrasing or common variations a user would say; it stays at 2 rather than 3 because coverage of natural terms is thin.

2 / 3

Distinctiveness Conflict Risk

Naming Blackbox and the multi-model judge gives it a niche, but 'coding tasks' overlaps with related skills like claude-code and codex, so it is only somewhat specific rather than clearly distinct (3).

2 / 3

Total

9

/

12

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.

Validation13 / 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.