CtrlK
BlogDocsLog inGet started
Tessl Logo
SkillPublic

claude-code-bridge

Integrate with Claude Code CLI for delegating complex coding tasks. Use when you need to spawn a separate Claude Code instance to work on files, run tests, or perform multi-step coding operations in a specific directory. Enables "Claude calling Claude" patterns for parallel work or specialized contexts.

Quality

81%

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

The canonical home for this skill is claude-code-bridge in Raudbjorn/TTTRPS

SKILL.md
Quality
Evals
Security

Quality

Content

75%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-organized, actionable single-file skill: concrete CLI commands, named MCP tools, and example workflows with implicit feedback loops. The main gaps are stylized non-executable MCP call syntax, verification stated as advice rather than workflow steps, and slight redundancy between the description and the body's opening sections.

Suggestions

Replace the stylized [claude_prompt] blocks with a literal, executable invocation example (actual tool-call arguments or CLI flags) so the MCP path is as copy-paste-ready as the CLI path.

Fold 'Check Results' into each example workflow as an explicit validation step (review output, then proceed/continue) to upgrade the implicit feedback loop to an explicit one.

Trim the intro sentence and When-to-Use list that restate the frontmatter description, or move the MCP configuration JSON blocks to a reference file to reduce inline tokens.

DimensionReasoningScore

Conciseness

The body is efficient: terse best practices ("Capture Session IDs: Save session IDs from responses to resume conversations"), compact error handling, and code examples with no explanation of concepts Claude already knows. It falls short of anchor 5 because the intro line "Delegate coding tasks to Claude Code CLI from any context" and the When-to-Use list largely restate the frontmatter description, and the three example workflows add tokens with marginal new information. It is above anchor 3 because there is no real over-explanation to strip.

4 / 5

Actionability

Mostly executable: concrete CLI commands with flags ("claude -p \"List all TODO comments in this project\" --output-format json", --continue, --resume <session-id>), named MCP tools, and complete MCP config JSON blocks. It is not anchor 5 because the MCP invocation examples use stylized pseudo-syntax ([claude_prompt] blocks with prompt:/working_dir: fields) rather than copy-paste-ready tool calls, and the Rust example depends on an unverified claude_code_rs crate. It is well above anchor 3 since the CLI path is fully executable.

4 / 5

Workflow Clarity

Example workflows give clear sequences, including a two-step followup pattern (identify deprecated API uses, then "Now migrate the first 3 files you found to API v2" after reviewing results), and Best Practices embed checkpoints like "Check Results: Verify Claude Code's output before proceeding" and "Capture Session IDs". Not anchor 5 because verification is stated as a bullet rather than an explicit validate-before-proceeding step woven into each workflow; not anchor 3 because checkpoints and sequences are present, only implicitly ordered.

4 / 5

Progressive Disclosure

Good structure with well-labeled sections (When to Use, Usage Patterns, Configuration, Best Practices, Example Workflows, Error Handling) and no dead or nested references — appropriate for a single-file skill with no bundle. It misses anchor 5 because at ~130 lines it exceeds the simple-skill threshold and content such as the two host-specific MCP config blocks and the example workflows could plausibly live in reference files; it is above anchor 3 because nothing is buried and navigation is easy.

4 / 5

Total

16

/

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 what-and-when structure and concrete trigger conditions. Its main deductions are the second-person "you need" phrasing (which costs it a specificity point under the rubric) and modest overlap risk with general agent-delegation skills.

Suggestions

Rewrite in third person to avoid the specificity penalty, e.g., "Integrates with the Claude Code CLI to delegate complex coding tasks. Use when a separate Claude Code instance is needed to work on files, run tests, or perform multi-step coding operations in a specific directory."

Add natural synonyms users might say, such as "subagent", "headless mode", or "background Claude session", to strengthen trigger-term coverage toward anchor 5.

Sharpen the trailing sentence "Enables 'Claude calling Claude' patterns for parallel work or specialized contexts" — 'specialized contexts' is vague and adds overlap risk with general delegation skills.

DimensionReasoningScore

Specificity

Base fit is anchor 4 — "delegating complex coding tasks" plus "work on files, run tests, or perform multi-step coding operations in a specific directory" lists several concrete actions with minor gaps. However, the second-person phrasing "Use when you need to spawn a separate Claude Code instance" violates the third-person-voice guideline, incurring the mandated 1-point penalty, so 4 - 1 = 3. It is not a 2 because the actions named are concrete and domain-anchored, not generic.

3 / 5

Completeness

Both halves are explicit: the what is "Integrate with Claude Code CLI for delegating complex coding tasks" and the when is the concrete trigger clause "Use when you need to spawn a separate Claude Code instance to work on files, run tests, or perform multi-step coding operations in a specific directory" — matching the anchor-5 pattern of clear what AND when with concrete trigger phrases. Not a 4 because the when-clause enumerates specific triggering conditions rather than a vague 'when working with X'.

5 / 5

Trigger Term Quality

Natural terms users would say are present: "delegating complex coding tasks", "spawn a separate Claude Code instance", "run tests", "parallel work", "Claude calling Claude". A few natural variations are missing (e.g., "subagent", "background task", "headless", "another Claude session"), which keeps it below anchor 5 but comfortably above anchor 3's partial keyword coverage.

4 / 5

Distinctiveness Conflict Risk

It carves a clear niche — spawning a separate Claude Code CLI instance, "Claude calling Claude" patterns — with distinct triggers, so it is well above anchor 3. It stays at 4 rather than 5 because phrases like "parallel work" and delegating coding tasks to another agent overlap with general subagent/Agent-tool skills, creating minor conflict risk with closely related delegation skills.

4 / 5

Total

16

/

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

Raudbjorn/TTTTRPS

Path

planning/done/claude-code-integration/claude-code-bridge/skill/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.