session-execution

Use when working on or reviewing session execution, command handling, shell state, FIFO-based streaming, or stdout/stderr separation. Relevant for session.ts, command handlers, exec/execStream, or anything involving shell process management. (project)

Quality

71%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

Securityby

Passed

No findings from the security scan

Fix and improve this skill with Tessl

tessl review fix ./.agents/skills/session-execution/SKILL.md

Quality

Content

82%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 tight, high-signal knowledge skill: project-specific architecture facts, concrete review checklists, and useful false-positive guidance for race conditions, all delivered with excellent token efficiency. Its only weaknesses are minor — deferred detail in repo docs that are not part of the bundle, and no explicit validation/retry loop for the suggested tests.

Dimension	Reasoning	Score
Conciseness	Lean bullet-based body with zero padding and no explanation of concepts Claude already knows — even the rationale lines ('bash waits for redirects to complete', 'robust on tmpfs/overlayfs') carry genuinely project-specific information. Every token earns its place.	5 / 5
Actionability	Concrete, checkable guidance throughout ('Verify exit code handling is atomic (write to .tmp then mv)', 'Check FIFO cleanup in error paths', 'Test silent commands (cd, variable assignment)'). Minor gaps: no commands or examples showing how to run the suggested tests, and the deep detail is deferred to docs rather than shown.	4 / 5
Workflow Clarity	The review checklist and the numbered race-condition triage procedure (same-session → cross-session → refer to CONCURRENCY.md) give a clear sequence, and the common-false-positive/actual-concern lists act as checkpoints. Minor gaps: the development section lists tests to consider but no validate-and-retry loop.	4 / 5
Progressive Disclosure	Well-organized sections with one-level-deep, clearly signaled references ('Read docs/SESSION_EXECUTION.md before working in this area', 'Refer to docs/CONCURRENCY.md for the full concurrency model'). The body runs slightly over 50 lines and the referenced docs live in the repo rather than the bundle, so it is good-not-perfect structure.	4 / 5
	Total	17 / 20 Passed

Description

61%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 well-crafted trigger description with rich, natural domain vocabulary and low conflict risk, but it is purely a 'when' clause — it never states what the skill provides. Adding one sentence of concrete capabilities (architecture guidance for reliable command execution, review checklists for concurrency correctness) would raise completeness and specificity.

Suggestions

Add a leading 'what' clause, e.g. 'Provides architecture context and review checklists for reliable session command execution with stdout/stderr separation.' before the 'Use when' trigger.

Name the concrete actions the skill supports (reviewing exit-code atomicity, FIFO cleanup, race-condition triage) rather than only the situations where it applies.

Tighten the broad tail trigger 'anything involving shell process management' to specific surfaces to further reduce overlap risk.

Dimension	Reasoning	Score
Specificity	Names the domain and concrete artifacts ('FIFO-based streaming', 'stdout/stderr separation', 'session.ts', 'exec/execStream') but the only actions stated are the generic 'working on or reviewing' — no specific capability statements. Fits anchor 3 (domain plus 1-2 concrete items, not comprehensive); not 4 because no concrete actions are listed.	3 / 5
Completeness	Only the 'when' is present; there is no 'what this skill does' statement anywhere, which matches the anchor-2 shape ('only when present without what'). The unusually detailed, multi-faceted trigger clause lifts it between anchors to 3; it cannot reach 4 because a 'what' is entirely absent.	3 / 5
Trigger Term Quality	Strong natural vocabulary — 'command handling', 'shell state', 'stdout/stderr separation', 'session.ts', 'exec/execStream', 'shell process management' — the phrases a developer would actually say. A few common variations are missing (e.g. 'terminal', 'background commands', 'streaming output'), so it falls just short of comprehensive anchor 5.	4 / 5
Distinctiveness Conflict Risk	Clear project niche with distinct triggers (session.ts, exec/execStream, FIFO streaming) that are unlikely to fire the wrong skill. The trailing 'anything involving shell process management' is somewhat broad, creating minor overlap risk, so 4 rather than 5.	4 / 5
	Total	14 / 20 Passed

Validation

100%

Warnings & errors only

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: cloudflare/sandbox-sdk
Path: .agents/skills/session-execution/SKILL.md
Commit: 831b8c5

Reviewed: 7 days ago

Table of Contents

Discovery Implementation Validation

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.