CtrlK
BlogDocsLog inGet started
Tessl Logo

ccb-config

Private built-in CCB configuration skill for agentroles.ccb_self. Design, edit, validate, and prepare reloads for .ccb/ccb.config, role bindings, providers, windows, workspaces, tool windows, sidebar, and provider startup inputs. Use only inside ccb_self; non-self agents should delegate CCB config changes to ccb_self.

70

Quality

86%

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

SKILL.md
Quality
Evals
Security

Quality

Content

88%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 well-built operational skill: fully executable commands, a gated 12-step workflow with validation and rollback feedback loops, and a real one-level-deep reference. The main improvement areas are trimming the duplication between the Scope and Required Workflow sections and tightening the overlap between the body's Role Binding details and references/config-contracts.md.

Suggestions

Collapse the duplicated validate/dry-run/reload gates in the Scope section into the Required Workflow, keeping Scope as a short allowed/forbidden list.

Move the detailed Role Binding TOML examples and overlay rules into references/config-contracts.md (or add a labeled "See references/config-contracts.md" section) so the body stays a true overview.

Turn the inline "Read references/config-contracts.md..." sentence into a clearly labeled reference section so navigation is explicit.

DimensionReasoningScore

Conciseness

The body is dense and rule-driven with no concept explanations Claude already knows, giving exact commands, TOML snippets, and a handoff template. It fits 'efficient; minor instances that could be trimmed': the validate/dry-run/reload gates appear in both the Scope section ("Run ccb config validate after every edit", "Run ccb reload --dry-run before reload materialization") and again in Required Workflow steps 6-10, a modest redundancy that keeps it below the every-token-earns-its-place anchor at 5 and well above the verbose anchors.

4 / 5

Actionability

Guidance is fully executable and copy-paste ready: "ccb config validate", "ccb reload --dry-run", "cp .ccb/ccb.config .ccb/ccb.config.bak.$(date +%s)", "ccb roles install agentroles.ccb_self", plus concrete TOML binding examples and a literal handoff block. This matches the top anchor for executable commands covering the common cases; there is no vague or pseudocode direction that would drop it to 4.

5 / 5

Workflow Clarity

The 12-step Required Workflow has a clear sequence with explicit validation checkpoints (validate after every edit, dry-run before materialization) and a genuine error-recovery feedback loop in step 7 ("report the full validation error, do not run reload... Restore the previous config when a reliable pre-edit copy exists; otherwise stop and ask"). This is exactly the validate → fix/rollback → retry pattern the top anchor describes; the batch/destructive-operation cap at 3 does not apply because validation is pervasive.

5 / 5

Progressive Disclosure

The body points to a real, one-level-deep reference with a clear condition — "Read `references/config-contracts.md` before complex edits or reload-impact analysis" — and the file exists with appropriate depth. It sits at the 'good structure; minor organization gaps' anchor rather than 5 because the pointer is an inline sentence rather than a labeled navigation section, and the Role Binding TOML rules in the body substantially duplicate the reference's Role Bindings and Windows Topology sections, so the split between overview and reference could be cleaner.

4 / 5

Total

18

/

20

Passed

Description

83%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: concrete, comprehensive action coverage in third person, with explicit scoping that makes it highly distinctive. Its main weakness is the 'when' side — usage guidance is framed as a scope restriction ("Use only inside ccb_self") rather than natural task-trigger phrases, and a few obvious trigger terms are absent.

Suggestions

Add an explicit trigger phrase such as "Use when the user asks to change, validate, or reload .ccb/ccb.config, agent windows, or provider settings" to sharpen the 'when' half.

Include natural user-facing variations like "ccb reload", "ccb setup", and "agent roles" so the skill matches how requests are actually phrased.

DimensionReasoningScore

Specificity

The description lists multiple concrete actions — "Design, edit, validate, and prepare reloads for .ccb/ccb.config, role bindings, providers, windows, workspaces, tool windows, sidebar, and provider startup inputs" — with comprehensive coverage of the config surface. It matches the anchor for multiple specific concrete actions and comprehensive coverage; nothing is generic or padded, so it does not fall to the 4 anchor with coverage gaps.

5 / 5

Completeness

The 'what' is clear and concrete (design/edit/validate/prepare reloads for the enumerated config surfaces), and a 'when' is present via "Use only inside ccb_self; non-self agents should delegate CCB config changes to ccb_self" — explicit trigger guidance, so the missing-'Use when' cap at 3 does not apply. However, the 'when' covers scope restrictions rather than natural task triggers, matching the anchor where 'when' could be more explicit or specific rather than the fully explicit trigger phrases at 5.

4 / 5

Trigger Term Quality

It contains good, natural keywords — "ccb.config", "reload", "role bindings", "providers", "validate" — but is missing common variations a user might say ("ccb reload", "agent roles", "ccb setup"). This fits the 'good keyword coverage; a few natural terms missing' anchor rather than the comprehensive-synonyms anchor at 5.

4 / 5

Distinctiveness Conflict Risk

It carves a clear niche: "Private built-in CCB configuration skill for agentroles.ccb_self", with an explicit scoping instruction that non-self agents must delegate. Combined with the unique `.ccb/ccb.config` file path, conflict risk with other skills is minimal, matching the top anchor. It is not merely 'mostly distinct' as at 4 — the delegation clause actively prevents mis-triggering.

5 / 5

Total

18

/

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
SeemSeam/claude_codex_bridge
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.